Study Of Erroranal Ysisand Kinematic Calibration Of New Five-Dof Polishing Parallelkinema Ticsmachine

Parallel Kinematics Machine (PKM) is a creative application of researchachievements of advanced spatial mechanism in the field of NC machine tool. It hasmany superiorities to traditional machine tool such as good dynamic performance,keeping high rigidity and high precision in case of many degrees of freedom.PKM has beapplied excellently in mechanical manufacturing field. But the bottleneck of applicationof PKM is still the low acrrucy and the error difficult to control.The mainly error of PKM is due to quasi-static error and dynamic error, and 70%error of PKM is the quasi-static error. And geometric error factors are the major of thequasi-static error. Then, to improve the acrracy error of PKM , the work about geometricacrracy design and the calibration of geometric error should be done. Aiming at the needof practice, this paper is focused on a new polishing parallel kinematics machine of five-DOF and 3T2R from the above two aspects. The mainly containing the following aspects:Based on the theory of robot machanism, against the PKM studied by this paper, theinverse and forward kinematic analysis of the new polishing parallel kinematicsmachine of five-DOF is formulated. .And laid the foundation for the follow-up work ofaccuracy analysis and kinematic calibration.According to the measuring characteristics of the barball, considering the constructof the PKM studied by this paper,the error model and the geometrical parameter identificationmodel of the PKM were established by the space vector looped method and f theerror analysis work with the method of orthogonal experimental design was done toprovide the theoretical basis for geometric acrracy design.Lastly,the proving program was put forward for the method of calculating machineerrors and kinematic calibration, based on emulation method with Matlab. Through thisprocess, the error model and the geometrical parameter identification model of the PKMestablished by this paper are proved to be validated. Then the controlling precision can beimproved by the models established in this paper.