Research On Chatter Analysis And Machining Accuracy Improvement Method Of Industrial Robotic Milling System

As a critical important equipment in industrial scenes,the improvement and application of industrial robots and their control technologies are important means and key links to promote the development of Chinese intelligent manufacturing.In recent years,industrial robots are widely used in automatic production,electronic appliances,food,chemical industry,aerospace and other industries due to their superiority of small space occupation,low cost and high flexibility.The applied processing involved include welding,handling,sorting,drilling,grinding,polishing,etc.Different from the drilling,grinding,polishing and other processes,the problem of milling process is harder.The workpiece is in rigid contact with the cutting tool of the robot,the cutting load is large,and the dynamic characteristics of the cutting process are complex,leading to unstable system and low accuracy.How to solve the conflict between low precision and weak stiffness of robots and high precision and high efficiency of machining requirements is a research focus of scholars at home and abroad.This dissertation focuses on the chatter analysis and the optimization method of machining accuracy of robotic milling system.First of all,in order to control the impact of chatter on the robot machining system and machining quality within an acceptable range,by studying the chatter detection method in robotic milling,the chatter phenomenon in the robot milling process can be found early and the damage caused by the uncontrolled chatter can be avoided.Then,by establishing a dynamic model of the robotic milling system considering both cross-modal coupling and regeneration effect,the chatter stability region of robotic milling is explored,helping experimenter to choose stable and chatter-free parameters.Then,in the case of no-load robot,analyze the source of robot positioning error,study the prediction and compensation method of robot positioning error,improve the absolute positioning accuracy of the no-load robot,and provide the foundation for improving the milling accuracy of the robot.Finally,in the case of robotic milling,by constructing a deformation model of the milling robot,a comprehensive deformation index of the robot milling process is proposed,and the robot milling posture is optimized based on this index to reduce the flexible deformation of the robot caused by the cutting load,thereby improving the robot machining precision.The main research contents and implementation plans are summarized as below:(1)A six-degree-of-freedom industrial robot milling experiment platform was built,and the robot motion control-milling integrated control system software was developed.By analyzing the requirements of the industrial robot milling experiment platform,the robot function requirements,milling function requirements and measurement requirements are determined,and the hardware composition scheme of the robot platform of milling is formulated.Then,based on the standard PC,the robot motion control-milling integrated control system software is developed,which realizes the synergistic control of robot motion and milling,and provides a basis for the robot milling experiment and the subsequent verification of the robot chatter analysis and machining accuracy optimization algorithm.(2)The extraction of chatter features in robot vibration signals is studied,and an on-line detection method for robot milling chatter based on variational modal decomposition algorithm is proposed.Firstly,based on the instantaneous frequency and kurtosis coefficient,an adaptive variational modal decomposition method for vibration signals is proposed,which realizes the adaptive solution of the upper limit of modal components and avoids the occurrence of modal mixing in the decomposition process.Then,genetic algorithm is used to achieve the optimization of decomposition parameters,including the number of modal components and penalty factor,and the selection strategy of the discrete step size of penalty factor is given according to the optimization result,which improves the calculation efficiency.Finally,a chatter feature based on the variation trend of entropy value is defined:entropy drift coefficient.The entropy drift coefficient based on approximate entropy and energy entropy realizes the detection of chatter in robotic milling.The simulation and experimental results verify that the the real-time requirements of online detection proposed chatter detection method can meet,and can effectively detect the occurrence of chatter in the milling process.(3)The basic principles and methods of chatter stability analysis of robotic milling system are clarified,and the experimental exploration of the chatter stability domain of robotic milling are carried out.First,a dynamic model of the robotic milling system considering the crossmodal coupling is established,and the differential equation of the system dynamics under regeneration effect is deduced accordingly.Then,the modal space dynamics model is constructed based on the asymmetric frequency response function model,and the hammering experiment is carried out on the robot milling processing system,and the system frequency response function curve is obtained by calculation.Then,a modal parameter identification method based on particle swarm optimization is proposed,and the modal parameters are obtained through the overall fitting of the frequency response function curve.Finally,the semidiscrete method is used to solve the chatter stability region of robotic milling.The experimental results show that the proposed modal parameter identification method can effectively improve the fitting accuracy of the frequency response curve.The results obtained from the stability prediction based on the identified modal parameters have certain accuracy,which can provide a theoretical guidance for selecting stable and chatter-free machining parameters.(4)The source of the positioning error and the classification problem under the no-load condition of the robot are studied,and an error compensation method based on the stacking model integration strategy is proposed.Firstly,based on the kinematic model and structure of the industrial robot,the source of the positioning error of the industrial robot is analyzed,and the positioning error compensation scheme is formulated by classifying the source of the error.Then,according to the robot milling workspace,the Latin hypercube sampling method is used to generate the error sampling data position points covering the workspace.Finally,the robot positioning error prediction model is established based on the stacking model integration strategy,and the mod based on the error data to realize the prediction and compensation of the end positioning error.The experimental results show that the proposed error prediction method can accurately predict the spatial position error of the robot end,and the positioning accuracy of the robot after compensation has been greatly improved.(5)The flexible deformation model of the robot milling process is studied,and the robot machining accuracy optimization method based on the comprehensive deformation index is proposed.Based on the static stiffness model of the robot,the deformation model of robot milling is constructed by analyzing the source of flexible deformation in the milling process,and a comprehensive deformation index of robot milling is proposed.Then,by taking the redundant degree of freedom as control variable,setting deformation index and singularity index as optimization goals,combined with joint angle and angular velocity constraints,the robot milling attitude optimization model can be constructed.Then,in order to ensure the enhancement of the robot’s stiffness performance,an optimization method within the ideal stiffness interval is proposed,and an adaptive genetic algorithm is used to optimize the robot’s robotic milling posture.The simulation and experimental results show that the proposed comprehensive deformation index can accurately reflect the deformation of the robot end,and optimizing the robot posture based on the deformation index can effectively reduce the deformation of the robot end and improve the milling accuracy.