Error Source Analysis And Modeling Of Self-driven AACMM

With the demand of intelligent manufacturing for on-line automatic measurement index of measuring tools,the traditional articulated arm coordinate measuring machine can not meet the requirements of intelligent manufacturing due to its low efficiency,uncertain measuring posture,low measuring accuracy and poor reproducibility of results due to its manual measurement work mode.Based on the research results of traditional manual joint arm coordinate measuring machine and series robot,the research and development of self-driven AACMM is under way.Since the self-drvien working mode had dynamic and static error sources in the measurement process,the measuring accuracy of the measuring machine is affected.In order to analyze the source of error,the working principle of the measuring machine was studied and analyzed in this paper,and the kinematic model of the measuring machine was established by combining the mechanical structure dimensions of the measuring machine.According to the structure design of the measuring machine,the range of rotation angle of each joint was determined and the working range of the measuring machine was solved.At the same time,the error was divided into positioning error and measurement error according to the measuring process of the measuring machine.Through theoretical analysis and calculation of each error source item,the influence value of each item was determined.Each error item was transformed into kinematics model parameters of the measuring machine,and error correction table of kinematics model parameters was listed,and single error partial differential model of the measuring machine was established.In order to verify the correctness of theoretical calculation of error source term of measuring machine,the results of theoretical calculation was taken as the boundary condition of simulation in this paper,and the error items which have great influence on the measuring machine were simulated and analyzed by ansys.Finally,each error was compared and validated,the main source terms of positioning error and measurement error were determined,and the full differential error model of the measuring machine was established.In order to get the predictive model of measuring machine probe position e rror,the angular velocity and angle value of each joint were determined as the main parameters affecting the dynamic error of the measuring machine by theor etical analysis.According to the results of the previous analysis,the joint parts that bring the main error source term were processed with flexible parameters o ne by one,The rigid-flexible coupling simulation model was built in adams.Th en the simulation analysis was carried out by using the trajectory of quintic pol ynomial programming,and the simulation data of the position error and motion parameters of the measuring machine were obtained.The simulated data was us ed as the training sample of the neural network.Through training and testing,t he prediction model of probe position error of the measuring machine was estab lished.The error source analysis and the establishment of the total differential error model of the self-driven AACMM are completed,through the analysis of the working principle and structure of the self-driven AACMM in this paper.and the theoretical calculation error value is verified by simulation.Finally,an error prediction model with 0.0988mm accuracy is established by using RBF neural network.Figure[62]Table[27]Reference[83]...