A multi-modal and acute curvature path-oriented composite robot force control polishing process parameter optimization method

Abstract

The application belongs to the field of process parameter optimization, and discloses a composite material robot force control grinding and polishing process parameter optimization method for multi-modal and rapidly changing curvature paths. The method realizes resonance regulation based on stiffness-pose coupling modeling, constructs parameter redundancy tolerance of multi-source uncertainty disturbance, and plans path coordination considering dynamic response. Through deep coupling of structure modal analysis, robot dynamic characteristics and robust design, a "workpiece-robot" system dynamics model and path look-ahead mechanism are constructed. The method changes the traditional experience-driven mode to an intelligent control mode, significantly improves the machining stability, material removal uniformity and overall quality of complex parts such as composite material blades under rapidly changing curvature, and realizes the optimization and collaborative control of the whole process parameters.

Description

Technical Field

[0001] This invention belongs to, but is not limited to, the field of process parameter optimization, and particularly relates to a method for optimizing the process parameters of force-controlled grinding and polishing of composite materials for multimodal and rapidly changing curvature paths. Background Technology

[0002] Aero-engines are core power equipment in the defense and civil aviation sectors. Improving their core performance, such as thrust-to-weight ratio and fuel efficiency, heavily relies on material upgrades and manufacturing precision assurance for fan / compressor blades. Fiber-reinforced composite blades offer advantages such as lightweight design, high specific strength, and excellent fatigue resistance. Their density is only one-third that of traditional high-temperature alloy blades, and their fatigue life can be more than ten times that of metal blades, making them a core component of next-generation high thrust-to-weight ratio aero-engines. After the composite blades are formed, their surface shape and position accuracy and contour roughness directly determine the engine's aerodynamic performance and service reliability. High-precision grinding and polishing are essential for achieving final shape control. Robotic grinding and polishing, with its high operational flexibility and strong adaptability to complex curved surfaces, has become the mainstream technology for precision grinding and polishing of composite blades.

[0003] However, the inherent material and geometric properties of composite blades present multiple technical challenges to optimizing robotic polishing process parameters. In terms of material properties, the laminated structure of composite materials results in significant anisotropy in the workpiece. The local stiffness and modal characteristics of different regions of the blade exhibit strong non-uniformity, easily triggering local modal resonance during polishing and causing processing defects such as drastic fluctuations in contact force, fiber pull-out, and matrix delamination. In terms of geometry, the airfoil surface of the blade has numerous abrupt curvature paths such as inlet and outlet edges. Limited robot acceleration and deceleration capabilities easily lead to instantaneous fluctuations in feed rate and contact compressive stress, causing non-uniform abrupt changes in material removal. As the core means of controlling polishing material removal behavior and suppressing processing disturbances, the rationality of the optimized design of process parameters directly determines the final processing quality of the composite blade.

[0004] Existing research and technical solutions for optimizing grinding and polishing process parameters of composite materials still suffer from the following insurmountable technical shortcomings:

[0005] The inability to adapt to the multimodal characteristics and non-uniform stiffness fields of composite materials easily leads to processing resonance and uneven material removal. Existing process parameter optimization is mostly aimed at homogeneous metal materials, and generally adopts a strategy of adjusting fixed parameters across the entire domain or a single force-controlled parameter. It does not consider the non-uniform distribution of local stiffness and multimodal characteristics caused by the ply structure of composite materials, and cannot establish a quantitative correlation between process parameters and workpiece vibration response. During grinding and polishing, local modes are easily excited in weak rigidity areas, causing structural resonance, resulting in excessive fluctuations in contact force, over-cutting / under-cutting of material removal, and difficulty in meeting the requirements of aerospace-grade machining for blade shape and position accuracy and surface consistency.

[0006] The lack of tolerance design for multi-source uncertainties leads to insufficient engineering robustness of the process scheme. The grinding and polishing process of composite materials involves multi-source uncertainties such as abrupt changes in path curvature, measured errors in workpiece stiffness, lag in the dynamic response of robot actuators, and wear of grinding tools. Existing optimization methods do not conduct sensitivity analysis of process parameters to disturbances, nor do they establish safety boundaries and redundant tolerance ranges for parameter adjustments. This results in process parameters easily exceeding accuracy thresholds after disturbances during on-site processing, leading to low form and position error compliance rates. Furthermore, laboratory-optimized process schemes are difficult to directly adapt to the complex working conditions of engineering sites.

[0007] The lack of multi-parameter coordinated control under rapidly changing curvature paths leads to significant deterioration in machining quality. Existing process optimizations for rapidly changing curvature paths mostly adjust only the feed rate parameter, failing to incorporate normal contact force, spindle speed, and robot joint dynamics into the coordinated optimization framework. This fails to address the fluctuations in robot inertial torque and abrupt changes in contact state caused by curvature abrupt changes, resulting in abrupt deviations in material removal in critical areas such as blade inlet and outlet edges, and a significant deterioration in surface quality and contour accuracy.

[0008] In summary, existing technologies cannot simultaneously solve the coupling problems of multimodal resonance suppression, robustness assurance of multi-source disturbances, and coordinated control of multiple parameters of rapidly changing curvature paths in composite materials. They are insufficient to meet the high-precision and high-consistency processing requirements of composite material blades for next-generation aero-engines. Therefore, it is urgent to develop a method for coordinated optimization of process parameters throughout the entire process that has the capabilities of prediction, compensation, and dynamic programming. Summary of the Invention

[0009] To address the problems existing in the prior art, this invention provides a method for optimizing the process parameters of force-controlled grinding and polishing of composite materials for multimodal and rapidly changing curvature paths.

[0010] This invention is implemented as follows: a method for optimizing the force-controlled grinding and polishing process parameters of composite materials for multimodal and rapidly changing curvature paths, the method comprising:

[0011] S1: A method for modeling and controlling the multimodal properties of composite materials based on stiffness-pose coupling;

[0012] S2: Construction of redundancy tolerance for process parameters under multi-source uncertainty disturbances;

[0013] S3: Collaborative planning of process parameters for rapidly changing curvature paths considering robot dynamic response.

[0014] Furthermore, S1 includes the following steps:

[0015] S1.1 Discretization and Transfer Model Establishment of Local Stiffness Field in Workpiece-Robot Coupled System:

[0016] First, the three-dimensional digital model and plying process parameters of the composite blade to be processed are obtained, including plying direction, number of layers, single-layer thickness, and material elastic constants. A global local stiffness field distribution model of the workpiece is constructed using finite element analysis. The grinding and polishing path is discretized into a series of small surface elements, and the local static stiffness of each element is calculated using finite element modal analysis. The reciprocal of the displacement under unit normal force and the natural frequency ;

[0017] Then, a kinematic model is established based on the robot-end effector DH parameters to determine the robot end effector pose for each unit, and subsequently, the joint angle vectors are obtained. Obtain the robot joint stiffness matrix through static loading tests or manuals. ,in For the first The stiffness coefficients of each joint; the Cartesian stiffness matrix of the end effector is calculated using the stiffness mapping formula:

[0018]

[0019] in For the robot's Jacobian matrix; from Extract the stiffness component along the tool contact normal. Treating the local stiffness of the workpiece and the stiffness of the robot's end effector as a series elastic system, we obtain the first... Workpiece-robot coupling stiffness of each unit : ;

[0020] Each unit is simplified as a single-degree-of-freedom spring-mass-damping system, with an equivalent mass... Damping ratio The transfer function of the system, obtained through modal analysis or experimental identification, is:

[0021]

[0022] in The natural frequency of the coupled system, The damping coefficient is used to establish a dynamic mapping relationship between "grinding and polishing excitation - coupled system stiffness - vibration response": given the normal contact force. Its frequency components can be used to calculate the steady-state vibration amplitude. The final result includes the coordinates of each element and the coupling stiffness. Natural frequency And a database of function parameters, providing input for subsequent steps;

[0023] S1.2 Identification of the multivariate mapping relationship between process parameters and vibration response considering robot dynamic characteristics: using the element coupling stiffness obtained in S1.1 and core process parameter—normal contact force Spindle speed Tool linear velocity For input variables, vibration amplitude and response frequency To obtain the output indicators, an orthogonal experiment was designed, covering different stiffness ranges and process parameter levels. The experiment was conducted on a robotic grinding and polishing platform, and vibration acceleration signals were collected simultaneously and obtained through FFT transformation. and Simultaneously, an excitation spectrum model was established based on contact mechanics theory to supplement sample data: grinding and polishing excitation frequency. The functional relationship with process parameters can be expressed as follows: Combined with transfer function Predictable vibration response, expanding dataset;

[0024] Using experimental and theoretical data, a nonlinear mapping model is constructed: The Sobol exponent sensitivity analysis was used to quantify the effect of each input variable on the output. The main effects and interaction effects were analyzed to determine the influence weights of each process parameter.

[0025] Establish excitation frequency With the natural frequency of the coupled system Association; Define frequency avoidance degree: ,in This is the modal order coefficient, usually taken as 1, but can be taken as 2 or 3 if higher-order modes are easily excited; the upper limit of vibration amplitude is determined according to processing quality requirements. Determine the critical avoidance threshold Construct excitation frequency avoidance constraints:

[0026] This constraint will be used to suppress resonance risk in subsequent dynamic programming of process parameters;

[0027] S1.3 Gradually varying segment parameter control strategy based on robot response bandwidth: based on the global coupling stiffness obtained in S1.1 Distribution and robot pose stiffness Distribution, combined with the curvature variation characteristics of the grinding and polishing path, identifies high-risk processing areas where abrupt changes in stiffness and rapid changes in curvature overlap, i.e., stiffness gradients. With curvature For larger areas, gradual process parameter change buffer zones are set before and after these areas, with buffer lengths... Based on stiffness abrupt gradient and robot maximum response bandwidth Determined, satisfies: , ,in This refers to the variation in process parameters. Here, represents the feed rate, and represents the time of change. Ensure that the rate of parameter change does not exceed the servo system's response capability;

[0028] Mapping model based on S1.2 Within the gradually changing section , , Continuous adjustment is performed.

[0029] Furthermore, step S2 includes the following steps:

[0030] S2.1 Establishment of a sensitivity analysis model for process parameter disturbances: based on the curvature of the grinding and polishing path. Measured stiffness error of coupled system Dynamic response error of robot actuator and the amount of wear on the abrasive. As an input perturbation variable, material is used to remove depth deviation. and blade shape and position error As an output response indicator, a perturbation sensitivity analysis model is established using the global Sobol sensitivity analysis method. First, the probability distribution model of each input perturbation variable is determined. Then, a large number of samples are generated through Monte Carlo simulation, and the material removal model and the form and position error model are substituted to calculate the variance of the output response. Highly sensitive perturbation factors are screened out through analysis to provide a basis for the subsequent construction of tolerance intervals.

[0031] S2.2 Identification of the Boundary Influence of Key Process Parameters: Identification of normal contact force through Monte Carlo numerical simulation and orthogonal verification experiments. Spindle speed and feed rate The influence boundary of key process parameters on output response indicators; specifically, according to curvature Different curvature regions and coupling stiffness of segments Within the stiffness regions of different coupled system segments, various process parameters were varied, and the deviation in material removal depth was observed. and form and position errors The changes in these parameters determine the allowable range of variation for the process parameters that meet the machining accuracy requirements, thus obtaining the influence boundary. ,in Represents a specific process parameter;

[0032] S2.3 Construction of Redundancy Tolerance Range and Safety Constraints for Process Parameters: Based on the influence boundaries identified in S2.2, redundancy tolerance ranges for process parameters corresponding to different processing regions are constructed. Typically, a subset of the influencing boundary is selected to ensure a safety margin. During the dynamic scheduling phase of the grinding and polishing process parameters, a safety boundary constraint mechanism is introduced to perform boundary correction and tolerance assessment on the real-time compensated process parameters. That is, for the process parameters adjusted in real-time... It must meet the following requirements: If it exceeds the range, constrain it to the nearest boundary value.

[0033] Furthermore, step S3 includes the following steps:

[0034] S3.1 Establishment of the Material Removal Amount Equality Constraint Model: Based on the material removal model of composite materials, establish the equality constraint relationship between process parameters and material removal amount; material removal depth. This can usually be expressed as: in This is a comprehensive coefficient. The index is calibrated experimentally. Contact time, relative to feed rate Related to the length of the contact area; based on the amount of target material removed. For constraints, we get: in This is the length of the grinding and polishing contact area;

[0035] S3.2 Construction of dynamic constraint equations for the robot considering the dynamic characteristics of the coupled system: Real-time acquisition of robot end-effector pose and joint angular velocity. Joint angular acceleration and driving torque Combining the robot's rigid body dynamics model: in The inertia matrix, The matrix of Coriolis force and centrifugal force. For gravity, The external force at the end point includes grinding and polishing contact forces; the workpiece-robot coupling stiffness matrix established in S1.1 is introduced. diagonal element is Considering the elastic deformation of the coupled system :

[0036] These constraints ensure that the robot operates safely within its dynamic response capability range, avoiding position overshoot or vibration caused by stiffness coupling;

[0037] S3.3 Multi-parameter collaborative optimization solution and dynamic update based on coupled constraints: The material removal amount equality constraint of S3.1, the robot dynamic constraint of S3.2, and the process parameter redundancy tolerance interval constraint of S2.3 are incorporated into a unified multi-parameter collaborative optimization framework; with the blade surface processing consistency and grinding and polishing efficiency as dual optimization objectives, a multi-objective optimization problem is constructed:

[0038] ,in .

[0039] Furthermore, in step S1.1, the static stiffness of each mesh element is obtained through finite element modal analysis. With natural frequency Solve the Cartesian stiffness matrix of the end effector under various postures using robot kinematics. and extract the normal stiffness The coupling stiffness is obtained by connecting the two in series. Simultaneously, the natural frequency of the workpiece corresponding to each unit is recorded. This will create a database corresponding to the grinding and polishing paths and robot configurations.

[0040] Furthermore, in step S1.3, the length of the gradually changing section... Based on stiffness abrupt gradient Determine the robot's maximum acceleration and deceleration performance to ensure that the rate of change of process parameters does not exceed the maximum response bandwidth of the servo system. ,Right now ,in .

[0041] Furthermore, in step S2.1, the global sensitivity analysis method adopts the Sobol sensitivity analysis method to quantify the main effect and interaction effect of each input perturbation variable on the material removal depth deviation and form and position error, and to screen out highly sensitive perturbation factors.

[0042] Furthermore, in step S3.3, the multi-objective intelligent optimization algorithm uses the non-dominated sorting genetic algorithm NSGA-II to obtain the Pareto optimal solution set, and selects the optimal combination of process parameters that takes into account both processing accuracy and processing efficiency from the optimal solution set based on the entropy weight method-TOPSIS method.

[0043] Based on the above technical solutions and the technical problems solved, the advantages and positive effects of the technical solution to be protected by this invention are as follows:

[0044] First, this invention constructs a full-process process parameter optimization method based on the dynamic characteristics of a "workpiece-robot" coupled system for grinding and polishing complex curved surfaces of composite materials. It realizes deep coupling and collaborative optimization between structural modal characteristic analysis, robot dynamic response characteristic analysis, robust design of process parameters, and path dynamic optimization. This transforms the grinding and polishing process from a traditional experience-driven mode to an intelligent control mode based on the dynamic model of the workpiece-robot coupled system and the path look-ahead planning mechanism, thereby significantly improving the processing stability, material removal uniformity, and overall processing quality of complex structural parts such as composite blades.

[0045] (1) Active suppression of resonance and control of material removal uniformity under the multimodal characteristics of composite materials were achieved. This invention establishes a nonlinear mapping relationship between process parameters and vibration response by constructing a coupled transmission model of "grinding and polishing excitation - local stiffness - vibration response". It innovatively proposes a parameter control strategy for the gradually changing section, which avoids structural resonance caused by local modal excitation in the weak rigid area of ​​composite blades from the root, effectively suppresses the fluctuation of grinding and polishing contact force, and controls the error of material removal in the whole blade area within ±3%. This solves the technical problem in the prior art that static process parameters cannot be adapted to the non-uniform stiffness field of composite materials, which easily leads to processing vibration and uneven removal.

[0046] (2) Significantly improves the robustness and engineering adaptability of grinding and polishing processes under multi-source uncertain disturbances. This invention establishes a disturbance sensitivity analysis model for the process parameter space, identifies the influence boundaries of key parameters, and constructs a regional process parameter redundancy tolerance interval and safety boundary constraint mechanism. It can effectively cope with multi-source uncertain disturbances such as sudden changes in path curvature, measured errors in workpiece stiffness, dynamic lag of actuators, and wear of grinding tools, thereby increasing the pass rate of blade shape and position error under disturbed conditions to over 98%, overcoming the shortcomings of existing process parameter optimization methods that lack disturbance tolerance design and have poor adaptability to engineering sites.

[0047] (3) This invention solves the problem of deteriorated processing quality caused by the lack of multi-parameter collaborative control under rapidly changing curvature paths. The invention deeply couples material removal characteristics with robot dynamic characteristics, and constructs a multi-parameter collaborative optimization framework that includes material removal equation constraints, robot dynamic constraints, and tolerance range constraints. It realizes the linkage optimization and dynamic update of normal contact force, feed rate, and spindle speed, and solves the problem that single parameter adjustment under rapidly changing curvature paths cannot take into account both processing accuracy and robot dynamic performance. This improves the surface processing consistency of rapidly changing curvature areas such as blade inlet and outlet edges by more than 40%, meeting the high-precision processing requirements of aerospace-grade composite material blades.

[0048] (4) It has complete engineering application value and technical promotion potential. The "modeling-tolerance-planning" full-process parameter optimization system formed by this invention can be deployed and implemented based on the existing industrial robot grinding and polishing platform without additional core hardware costs. The optimization logic is fully compatible with the existing robot CNC system and force control system. It is not only applicable to composite material blades of aero-engines, but can also be quickly promoted to robot grinding and polishing of various fiber-reinforced composite complex curved surface components such as wind turbine blades and aerospace thin-walled structural parts. It has significant economic benefits and engineering application value.

[0049] Secondly, as supporting evidence of the inventiveness of this invention, it is also reflected in the following important aspects:

[0050] (1) The technical solution of this invention fills a technical gap in the industry both domestically and internationally:

[0051] Existing composite material grinding and polishing process parameter optimization technologies mostly focus on optimizing a single influencing factor independently, and have not yet formed a comprehensive collaborative optimization system that can simultaneously address multimodal resonance suppression, robustness assurance of multi-source disturbances, and dynamic adaptation to rapidly changing curvature paths in composite materials. This invention, through a progressive design of three core steps, deeply integrates stiffness field-modal characteristics, multi-source disturbance uncertainties, robot dynamics characteristics, and process parameter optimization, constructing a systematic optimization scheme for the grinding and polishing process parameters of complex curved surfaces in composite materials. This fills the technical gap in the field of global collaborative optimization of process parameters under multi-factor coupled conditions.

[0052] (2) The technical solution of the present invention solves a technical problem that people have long wanted to solve but have never been able to solve successfully:

[0053] In the field of robotic grinding and polishing of composite blades, a core contradiction has long existed between "improving processing efficiency" and "ensuring dimensional and positional accuracy." High feed efficiency easily leads to dynamic fluctuations in regions with rapidly changing curvature and resonance in regions with weak rigidity, while conservative process parameters significantly reduce processing efficiency. This invention, through a combination of techniques including gradual-varying section control, redundancy tolerance constraints, and multi-parameter collaborative optimization, achieves precise control of overall processing accuracy and surface consistency while ensuring grinding and polishing efficiency, breaking through the long-standing technical bottleneck in the industry where efficiency and accuracy are difficult to balance.

[0054] (3) The technical solution of the present invention overcomes technical bias:

[0055] Traditional industry viewpoints generally hold that ensuring the dimensional and positional accuracy of composite material grinding and polishing hinges on improving the performance of the end effector, with process parameter optimization serving only as an auxiliary means. Furthermore, process parameter optimization for complex curved surfaces relies heavily on massive amounts of machining test data, making proactive control impossible through mechanistic modeling and forward-looking planning. This invention overturns this traditional understanding, demonstrating that through stiffness field-modal coupling modeling, disturbance tolerance design, and path forward-looking planning, machining vibrations can be proactively suppressed and machining errors compensated at the process level. Global optimization of process parameters can be achieved without relying on massive amounts of test data, overcoming the long-standing technical bias of "emphasizing hardware force control while neglecting proactive process optimization," and providing a completely new technical approach for the precision machining of composite materials. Attached Figure Description

[0056] Figure 1 This is a schematic diagram of the steps of a force-controlled grinding and polishing process parameter optimization method for composite materials robots oriented towards multimodal and rapidly changing curvature paths, provided by the present invention.

[0057] Figure 2 This is a schematic diagram of the path curvature and coupling stiffness distribution and high-risk area identification of the present invention;

[0058] Figure 3 This is a schematic diagram of the redundancy tolerance range of the process parameters in this invention;

[0059] Figure 4 This is a schematic diagram of the frequency avoidance constraint of the present invention;

[0060] Figure 5 This is a schematic diagram of the Pareto front and optimal solution selection for multi-objective optimization in this invention;

[0061] Figure 6 This is a physical image of a three-degree-of-freedom variable stiffness force control actuator mounted on a robotic grinding and polishing platform, as provided by this invention.

[0062] Figure 7 These are images showing the grinding and polishing results after applying the process parameters provided by this invention.

[0063] Figure 8 The image shows the result after grinding and polishing using the conventional process parameters provided by this invention. Detailed Implementation

[0064] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0065] This invention provides a method for optimizing the force-controlled grinding and polishing process parameters of composite materials using robots, oriented towards multimodal and rapidly changing curvature paths. Figure 1 As shown, it includes the following steps:

[0066] S1 A method for modeling and controlling the multimodal properties of composite materials based on stiffness-pose coupling;

[0067] S2 Construction of process parameter redundancy tolerance for multi-source uncertainty disturbances;

[0068] S3 considers collaborative planning of process parameters for rapidly changing curvature paths in robot dynamic response.

[0069] like Figure 1 As shown, step S1 includes the following steps:

[0070] S1.1 Discretization and Transfer Model Establishment of Local Stiffness Field in Workpiece-Robot Coupled System:

[0071] First, the three-dimensional digital model and plying process parameters (plying direction, number of layers, single-layer thickness, material elastic constants) of the composite blade to be processed are obtained. A global local stiffness field distribution model of the workpiece is then constructed using finite element analysis. The grinding and polishing path is discretized into a series of small surface elements, and the local static stiffness of each element is calculated using finite element modal analysis. (Reciprocal of displacement under unit normal force) and natural frequency .

[0072] Then, a kinematic model is established based on the robot-end effector DH parameters to determine the robot end effector pose for each unit, and subsequently, the joint angle vectors are obtained. Obtain the robot joint stiffness matrix through static loading tests or manuals. ,in For the first The stiffness coefficients of each joint. The Cartesian stiffness matrix of the end effector is calculated using the stiffness mapping formula:

[0073]

[0074] in For the robot's Jacobian matrix. From Extract the stiffness component along the tool contact normal. Treating the local stiffness of the workpiece and the stiffness of the robot's end effector as a series elastic system, we obtain the... Workpiece-robot coupling stiffness of each unit :

[0075] Each unit is simplified as a single-degree-of-freedom spring-mass-damping system, with an equivalent mass... Damping ratio The transfer function of the system, obtained through modal analysis or experimental identification, is:

[0076]

[0077] in The natural frequency of the coupled system, Let be the damping coefficient. Based on this, a dynamic mapping relationship is established between "grinding and polishing excitation—coupled system stiffness—vibration response": given the normal contact force... Its frequency components can be used to calculate the steady-state vibration amplitude. The final result includes the coordinates of each element and the coupling stiffness. Natural frequency And a database of passed function parameters, providing input for subsequent steps.

[0078] S1.2 Identification of the multivariate mapping relationship between process parameters and vibration response considering robot dynamic characteristics: using the element coupling stiffness obtained in S1.1 and core process parameter—normal contact force Spindle speed Tool linear velocity For input variables, vibration amplitude and response frequency To obtain the output indicators, an orthogonal experiment was designed, covering different stiffness ranges and process parameter levels. The experiment was conducted on a robotic grinding and polishing platform, and vibration acceleration signals were collected synchronously and obtained through FFT transformation. and Simultaneously, an excitation spectrum model was established based on contact mechanics theory, supplementing the sample data: grinding and polishing excitation frequency. The functional relationship with process parameters can be expressed as follows: Combined with transfer function It can predict vibration responses and expand the dataset.

[0079] Using experimental and theoretical data, a nonlinear mapping model is constructed: The Sobol exponent sensitivity analysis was used to quantify the effect of each input variable on the output. The main effects and interaction effects were analyzed to determine the influence weights of each process parameter.

[0080] Furthermore, establish incentive frequency With the natural frequency of the coupled system The correlation. Define the frequency avoidance degree: ,in This is the modal order coefficient (usually taken as 1, but can be 2 or 3 if higher-order modes are easily excited). It is determined based on processing quality requirements (upper limit of vibration amplitude). Determine the critical avoidance threshold Construct excitation frequency avoidance constraints:

[0081] This constraint will be used to suppress resonance risk in subsequent dynamic programming of process parameters, such as... Figure 2 As shown.

[0082] S1.3 Parameter control strategy for gradually varying sections based on robot response bandwidth: such as Figure 3 As shown, the global coupling stiffness obtained based on S1.1 Distribution and robot pose stiffness Distribution, combined with the curvature variation characteristics of the grinding and polishing path, identifies high-risk processing areas where abrupt changes in stiffness and rapid changes in curvature overlap, i.e., stiffness gradients. With curvature For larger areas, gradual process parameter change buffer zones are set before and after these zones, with buffer lengths... Based on stiffness abrupt gradient and robot maximum response bandwidth Determined, satisfies: , ,in This refers to the variation in process parameters. Here, represents the feed rate, and represents the time of change. Ensure that the rate of parameter change does not exceed the servo system's response capability.

[0083] Mapping model based on S1.2 Within the gradually changing section , , Continuous adjustment (such as linear interpolation or S-curve transition) is performed to adjust the grinding and polishing excitation frequency. Gradually deviating from the natural frequency of the coupled system This achieves gradual suppression of vibration response and avoids local modal resonance. After the control strategy is designed, the stiffness field information of the coupled system and the process parameter control strategy are integrated and embedded into the grinding and polishing execution system to achieve online adaptive control in high-risk areas.

[0084] like Figure 1 As shown, step S2 includes the following steps:

[0085] S2.1 Establishment of a sensitivity analysis model for process parameter disturbances: based on the curvature of the grinding and polishing path. Measured stiffness error of coupled system Dynamic response error of robot actuator and the amount of wear on the abrasive. As an input perturbation variable, material is used to remove depth deviation. and blade shape and position error As an output response indicator, a perturbation sensitivity analysis model was established using the global Sobol sensitivity analysis method. First, the probability distribution model of each input perturbation variable was determined. Then, a large number of samples were generated through Monte Carlo simulation, and substituted into the material removal model and the form and position error model to calculate the variance of the output response. Highly sensitive perturbation factors were identified through analysis, providing a basis for the subsequent construction of tolerance intervals.

[0086] S2.2 Identification of the Boundary Influence of Key Process Parameters: Identification of normal contact force through Monte Carlo numerical simulation and orthogonal verification experiments. Spindle speed and feed rate The influence boundary of key process parameters on output response indicators. Specifically, based on curvature... Different curvature regions and coupling stiffness of segments Within the stiffness regions of different coupled system segments, various process parameters were varied, and the deviation in material removal depth was observed. and form and position errors The changes in these parameters determine the allowable range of variation for the process parameters that meet the machining accuracy requirements, thus obtaining the influence boundary. ,in It represents a certain process parameter.

[0087] S2.3 Construction of Redundancy Tolerance Range and Safety Constraints for Process Parameters: Based on the influence boundaries identified in S2.2, redundancy tolerance ranges for process parameters corresponding to different processing regions are constructed. Typically, a subset of the influencing boundary is selected to ensure a safety margin. During the dynamic scheduling phase of the grinding and polishing process parameters, a safety boundary constraint mechanism is introduced to perform boundary correction and tolerance assessment on the real-time compensated process parameters, such as... Figure 4 As shown. That is, for the process parameters adjusted in real time. It must meet the following requirements: If the value exceeds the range, it is constrained to the nearest boundary value to ensure that the process parameters affected by the disturbance are always kept within the redundancy tolerance range, thereby improving the robustness of the processing.

[0088] like Figure 1 As shown, step S3 includes the following steps:

[0089] S3.1 Establishment of the Material Removal Amount Equation Constraint Model: Based on the material removal model for composite materials, an equation constraint relationship is established between process parameters and material removal amount. Material Removal Depth This can usually be expressed as: in This is a comprehensive coefficient. The index is calibrated experimentally. Contact time, relative to feed rate It is related to the length of the contact area. The amount of target material removed... For constraints, we get: in The length of the polishing contact zone can be simplified to a constant or related to curvature. This equation constraint will serve as a hard condition for subsequent optimization.

[0090] S3.2 Construction of dynamic constraint equations for the robot considering the dynamic characteristics of the coupled system: Real-time acquisition of robot end-effector pose and joint angular velocity. Joint angular acceleration and driving torque Combining the robot's rigid body dynamics model: in The inertia matrix, The matrix of Coriolis force and centrifugal force. For gravity, This represents the end-effector force (including grinding and polishing contact force). The workpiece-robot coupling stiffness matrix established in S1.1 is introduced. (Diagonal element is) Considering the elastic deformation of the coupled system This deformation will affect the actual position of the end effector, and thus the joint motion. Therefore, a dynamic constraint equation incorporating elastic deformation is constructed to limit the range of joint motion variation under abrupt curvature paths:

[0091] These constraints ensure that the robot operates safely within its dynamic response capability range, avoiding position overshoot or vibration caused by stiffness coupling.

[0092] S3.3 Multi-parameter Cooperative Optimization Solution and Dynamic Update Based on Coupled Constraints: The material removal amount equality constraint in S3.1, the robot dynamic constraint in S3.2, and the process parameter redundancy tolerance interval constraint in S2.3 are incorporated into a unified multi-parameter cooperative optimization framework. A multi-objective optimization problem is constructed with the dual optimization objectives of blade surface processing consistency (e.g., minimizing the standard deviation of removal depth at each point) and grinding and polishing efficiency (e.g., minimizing the total processing time):

[0093] ,in The non-dominated sorting genetic algorithm NSGA-II is used to solve for the Pareto optimal solution set. Then, the optimal combination of process parameters that balances machining accuracy and efficiency is selected from the optimal solution set based on the entropy weight method-TOPSIS. During the grinding and polishing process, the process parameters of the corresponding processing section are dynamically updated based on the real-time forward-looking detection results of the path curvature and the online identification results of the coupled system stiffness, so as to achieve stable processing under the path of rapid curvature change.

[0094] In the grinding and polishing path planning stage, a path curvature look-ahead window mechanism is introduced. By performing look-ahead analysis on the path segment that the robot is about to execute, the curvature change trend is identified in advance before entering the curvature change region. Combined with the stiffness distribution information of the workpiece-robot coupling system, the process parameters of the subsequent path segment are adjusted in advance, thereby avoiding transient dynamic impacts on the robot in the curvature change region.

[0095] During the grinding and polishing path planning stage, by jointly analyzing the path curvature change and the stiffness field distribution of the workpiece-robot coupling system, high-risk processing areas with curvature abrupt changes and low coupling stiffness regions are identified. In these areas, the gradual control of process parameters and multi-parameter collaborative optimization mechanism are triggered first to reduce the risk of structural vibration amplification and material removal anomalies.

[0096] In step S1.1, the static stiffness of each mesh element is obtained through finite element modal analysis. With natural frequency Solve the Cartesian stiffness matrix of the end effector under various postures using robot kinematics. and extract the normal stiffness The coupling stiffness is obtained by connecting the two in series. Simultaneously, the natural frequency of the workpiece corresponding to each unit is recorded. This will create a database corresponding to the grinding and polishing paths and robot configurations.

[0097] In step S1.3, the length of the gradually changing section Based on stiffness abrupt gradient Determine the robot's maximum acceleration and deceleration performance to ensure that the rate of change of process parameters does not exceed the maximum response bandwidth of the servo system. ,Right now ,in

[0098] In step S2.1, the global sensitivity analysis method adopts the Sobol sensitivity analysis method to quantify the main effect and interaction effect of each input perturbation variable on the material removal depth deviation and form and position error, thereby screening out highly sensitive perturbation factors.

[0099] In step S3.3, the multi-objective intelligent optimization algorithm uses the non-dominated sorting genetic algorithm NSGA-II to obtain the Pareto optimal solution set, and selects the optimal combination of process parameters that balances processing accuracy and processing efficiency from the optimal solution set based on the entropy weight method-TOPSIS method, such as... Figure 5 As shown.

[0100] Evidence related to the technical effects obtained by the embodiments of the present invention.

[0101] like Figure 6Based on a three-degree-of-freedom variable stiffness force-controlled actuator mounted on a robotic grinding and polishing platform, a systematic grinding and polishing comparison experiment was conducted using composite material blades as the processing object. Processing tests were performed using both a traditional fixed process parameter scheme and the optimized parameter generation method proposed in this paper, comprehensively comparing the surface morphology, roughness values, and overall processing consistency of the two groups of samples.

[0102] Experimental results show that the process parameter generation method proposed in this invention can combine the curved surface structure characteristics and material mechanical properties of composite blades to match the optimal grinding and polishing force, feed rate, and contact stiffness in real time, effectively avoiding defects such as over-grinding, under-grinding, material chipping, and surface fiber damage caused by rigid processing. Compared with traditional processing methods, the surface texture of the blade is more uniform and regular, and processing defects such as tool marks and scratches are significantly reduced; the surface roughness index of the workpiece is significantly optimized, and the surface finish and processing uniformity are greatly improved. At the same time, the adaptive adjustment capability of variable stiffness force control effectively reduces the contact impact during the grinding process of complex curved surfaces, improves the overall forming quality of composite blades after grinding and polishing, and verifies the feasibility and engineering practicality of this process parameter optimization method in the precision grinding and polishing scenario of complex components. The process parameters generated by the method of this invention result in the following after grinding and polishing: Figure 7 After grinding and polishing with conventional process parameters, as follows: Figure 8 .

[0103] It should be noted that embodiments of the present invention can be implemented in hardware, software, or a combination of both. The hardware portion can be implemented using dedicated logic; the software portion can be stored in memory and executed by a suitable instruction execution system, such as a microprocessor or dedicated-design hardware. Those skilled in the art will understand that the above-described devices and methods can be implemented using computer-executable instructions and / or included in processor control code, for example, such code provided on a carrier medium such as a disk, CD, or DVD-ROM, a programmable memory such as read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The devices and modules of the present invention can be implemented by hardware circuitry such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field-programmable gate arrays, programmable logic devices, etc., or by software executed by various types of processors, or by a combination of the above-described hardware circuitry and software, such as firmware.

[0104] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions, and improvements made by those skilled in the art within the scope of the technology disclosed in the present invention, and within the spirit and principles of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for optimizing the force-controlled grinding and polishing process parameters of composite materials using robots for multimodal and rapidly changing curvature paths, characterized in that, The method includes: S1: A method for modeling and controlling the multimodal properties of composite materials based on stiffness-pose coupling; S2: Construction of redundancy tolerance for process parameters under multi-source uncertainty disturbances; S3: Collaborative planning of process parameters for rapidly changing curvature paths considering robot dynamic response.

2. The method for optimizing the force-controlled grinding and polishing process parameters of composite materials robots for multimodal and rapidly changing curvature paths according to claim 1, characterized in that, S1 includes the following steps: S1.1 Discretization and Transfer Model Establishment of Local Stiffness Field in Workpiece-Robot Coupled System: First, the three-dimensional digital model and plying process parameters of the composite blade to be processed are obtained, including plying direction, number of layers, single-layer thickness, and material elastic constants. A global local stiffness field distribution model of the workpiece is constructed using finite element analysis. The grinding and polishing path is discretized into a series of small surface elements, and the local static stiffness of each element is calculated using finite element modal analysis. The reciprocal of the displacement under a unit normal force and the natural frequency ; Then, a kinematic model is established based on the robot-end effector DH parameters to determine the robot end effector pose for each unit, and subsequently, the joint angle vectors are obtained. Obtain the robot joint stiffness matrix through static loading tests or manuals. ,in For the first The stiffness coefficients of each joint; the Cartesian stiffness matrix of the end effector is calculated using the stiffness mapping formula: in For the robot's Jacobian matrix; from Extract the stiffness component along the tool contact normal. Treating the local stiffness of the workpiece and the stiffness of the robot's end effector as a series elastic system, we obtain the first... Workpiece-robot coupling stiffness of each unit : ; Each unit is simplified as a single-degree-of-freedom spring-mass-damping system, with an equivalent mass... Damping ratio The transfer function of the system, obtained through modal analysis or experimental identification, is: in The natural frequency of the coupled system, The damping coefficient is used to establish a dynamic mapping relationship between "grinding and polishing excitation - coupled system stiffness - vibration response": given the normal contact force. Its frequency components can be used to calculate the steady-state vibration amplitude. The final result includes the coordinates of each element and the coupling stiffness. Natural frequency And a database of function parameters, providing input for subsequent steps; S1.2 Identification of the multivariate mapping relationship between process parameters and vibration response considering robot dynamic characteristics: using the element coupling stiffness obtained in S1.1 and core process parameter—normal contact force Spindle speed Tool linear velocity For input variables, vibration amplitude and response frequency To obtain the output indicators, an orthogonal experiment was designed, covering different stiffness ranges and process parameter levels. The experiment was conducted on a robotic grinding and polishing platform, and vibration acceleration signals were collected simultaneously and obtained through FFT transformation. and Simultaneously, an excitation spectrum model was established based on contact mechanics theory to supplement sample data: grinding and polishing excitation frequency. The functional relationship with process parameters can be expressed as follows: Combined with transfer function Predictable vibration response, expanding dataset; Using experimental and theoretical data, a nonlinear mapping model is constructed: The Sobol exponent sensitivity analysis was used to quantify the effect of each input variable on the output. The main effects and interaction effects were analyzed to determine the influence weights of each process parameter. Establish excitation frequency With the natural frequency of the coupled system Association; Define frequency avoidance degree: ,in This is the modal order coefficient, usually taken as 1, but can be taken as 2 or 3 if higher-order modes are easily excited; the upper limit of vibration amplitude is determined according to processing quality requirements. Determine the critical avoidance threshold Construct excitation frequency avoidance constraints: This constraint will be used to suppress resonance risk in subsequent dynamic programming of process parameters; S1.3 Gradually varying segment parameter control strategy based on robot response bandwidth: based on the global coupling stiffness obtained in S1.1 Distribution and robot pose stiffness Distribution, combined with the curvature variation characteristics of the grinding and polishing path, identifies high-risk processing areas where abrupt changes in stiffness and rapid changes in curvature overlap, i.e., stiffness gradients. With curvature For larger areas, gradual process parameter change buffer zones are set before and after these areas, with buffer lengths... Based on stiffness abrupt gradient and robot maximum response bandwidth Confirmed, satisfied: , ,in This refers to the variation in process parameters. Here, represents the feed rate, and represents the time of change. Ensure that the rate of parameter change does not exceed the servo system's response capability; Mapping model based on S1.2 Within the gradually changing section , , Continuous adjustment is performed.

3. The method for optimizing the force-controlled grinding and polishing process parameters of composite materials robots for multimodal and rapidly changing curvature paths as described in claim 1, characterized in that, Step S2 includes the following steps: S2.1 Establishment of a sensitivity analysis model for process parameter disturbances: based on the curvature of the grinding and polishing path. Measured stiffness error of coupled system Dynamic response error of robot actuator and the amount of wear on the abrasive. As an input perturbation variable, material is used to remove depth deviation. and blade shape and position error As an output response indicator, a perturbation sensitivity analysis model is established using the global Sobol sensitivity analysis method. First, the probability distribution model of each input perturbation variable is determined. Then, a large number of samples are generated through Monte Carlo simulation, and the material removal model and the form and position error model are substituted to calculate the variance of the output response. Highly sensitive perturbation factors are screened out through analysis to provide a basis for the subsequent construction of tolerance intervals. S2.2 Identification of the Boundary Influence of Key Process Parameters: Identification of normal contact force through Monte Carlo numerical simulation and orthogonal verification experiments. Spindle speed and feed rate The influence boundary of key process parameters on output response indicators; in terms of curvature Different curvature regions and coupling stiffness of segments Within the stiffness regions of different coupled system segments, various process parameters were varied, and the deviation in material removal depth was observed. and form and position errors The changes in these parameters determine the allowable range of variation for the process parameters that meet the machining accuracy requirements, thus obtaining the influence boundary. ,in Represents a specific process parameter; S2.3 Construction of Redundancy Tolerance Range and Safety Constraints for Process Parameters: Based on the influence boundaries identified in S2.2, redundancy tolerance ranges for process parameters corresponding to different processing regions are constructed. Typically, a subset of the influencing boundary is selected to ensure a safety margin. During the dynamic scheduling phase of the grinding and polishing process parameters, a safety boundary constraint mechanism is introduced to perform boundary correction and tolerance assessment on the real-time compensated process parameters. That is, for the process parameters adjusted in real-time... It must meet the following requirements: If it exceeds the range, constrain it to the nearest boundary value.

4. The method for optimizing the force-controlled grinding and polishing process parameters of composite materials robots for multimodal and rapidly changing curvature paths according to claim 1, characterized in that, Step S3 includes the following steps: S3.1 Establishment of the Material Removal Amount Equality Constraint Model: Based on the material removal model of composite materials, establish the equality constraint relationship between process parameters and material removal amount; material removal depth. This can usually be expressed as: in This is a comprehensive coefficient. The index is calibrated experimentally. Contact time, relative to feed rate Related to the length of the contact area; based on the amount of target material removed. For constraints, we get: in This is the length of the grinding and polishing contact area; S3.2 Construction of dynamic constraint equations for the robot considering the dynamic characteristics of the coupled system: Real-time acquisition of robot end-effector pose and joint angular velocity. Joint angular acceleration and driving torque Combining the robot's rigid body dynamics model: in The inertia matrix, The matrix of Coriolis force and centrifugal force. For gravity, The external force at the end point includes grinding and polishing contact forces; the workpiece-robot coupling stiffness matrix established in S1.1 is introduced. diagonal element is Considering the elastic deformation of the coupled system : These constraints ensure that the robot operates safely within its dynamic response capability range, avoiding position overshoot or vibration caused by stiffness coupling; S3.3 Multi-parameter collaborative optimization solution and dynamic update based on coupled constraints: The material removal amount equality constraint of S3.1, the robot dynamic constraint of S3.2, and the process parameter redundancy tolerance interval constraint of S2.3 are incorporated into a unified multi-parameter collaborative optimization framework; with the blade surface processing consistency and grinding and polishing efficiency as dual optimization objectives, a multi-objective optimization problem is constructed: ,in .

5. The method for optimizing the force-controlled grinding and polishing process parameters of composite materials robots for multimodal and rapidly changing curvature paths according to claim 1, characterized in that, In step S1.1, the static stiffness of each mesh element is obtained through finite element modal analysis. With natural frequency Solve the Cartesian stiffness matrix of the end effector under various postures using robot kinematics. and extract the normal stiffness The coupling stiffness is obtained by connecting the two in series. Simultaneously, the natural frequency of the workpiece corresponding to each unit is recorded. This will create a database corresponding to the grinding and polishing paths and robot configurations.

6. The method for optimizing the force-controlled grinding and polishing process parameters of composite materials robots for multimodal and rapidly changing curvature paths according to claim 1, characterized in that, In step S1.3, the length of the gradually changing section Based on stiffness abrupt gradient Determine the robot's maximum acceleration and deceleration performance to ensure that the rate of change of process parameters does not exceed the maximum response bandwidth of the servo system. ,Right now ,in .

7. The method for optimizing the force-controlled grinding and polishing process parameters of composite materials robots for multimodal and rapidly changing curvature paths according to claim 1, characterized in that, In step S2.1, the global sensitivity analysis method adopts the Sobol sensitivity analysis method to quantify the main effect and interaction effect of each input perturbation variable on the material removal depth deviation and form and position error, and to screen out highly sensitive perturbation factors.

8. The method for optimizing the force-controlled grinding and polishing process parameters of composite materials robots for multimodal and rapidly changing curvature paths according to claim 1, characterized in that, In step S3.3, the multi-objective intelligent optimization algorithm uses the non-dominated sorting genetic algorithm NSGA-II to obtain the Pareto optimal solution set, and selects the optimal combination of process parameters that takes into account both processing accuracy and processing efficiency from the optimal solution set based on the entropy weight method-TOPSIS method.

Images