关键词: 并联机构, 优化设计, 运动学, 刚度, 动力学

Abstract: A three-level orderly proceeding optimization methodology was proposed based on kinematics， stiffness and dynamics of PM. The three intelligent algorithms of genetic algorithm（GA）， particle swarm algorithm（PSO） and differential evolution（DE） were applied to optimize the dimensional parameters， sectional parameters and mass parameters respectively. Firstly， optimization of the dimensional parameters was aiming at the workspace performance and motion/force transmission performance. Secondly， on the basis of the dimension parameter optimization， the sectional parameters of the components with the objective of the bearing stiffness and overall stiffness of the mechanisms were optimized. Finally， on the basis of the dimensional parameters and sectional parameter optimizations， the mass parameters of the mechanisms were optimized with the objective of dynamics dexterity and energy transmission efficiency. A zero-coupling SCARA PM was used as an example to illustrate the optimization procedure. Firstly， the kinematics model of the PM was established by using vector method， and the dimensional parameters were optimized through the workspace performance and motion/force transmission performance. Secondly， the stiffness model of the PM was obtained by using the virtual spring method， and the flexibility matrix was obtained through the parameter identification of CAD model. Then， the dynamics model of the PM was derived by using the general dynamics equation， and the mass parameters of the components were optimized through the dynamics dexterity and energy transmission efficiency. Thus， the comprehensive performances of kinematics， stiffness and dynamics of the PM reached. Meanwhile， some design criteria of dimensional， sectional and mass parameters were obtained. 

Key words: parallel mechanism(PM), optimization design, kinematics, stiffness, dynamics