摘要
为了控制装配过程中的关键装配特性,以大尺寸测量技术为辅助,实现大型零部件最优位姿装配,提出基于关键装配特性的大型零部件最佳装配位姿多目标优化算法。该方法将测量辅助装配(MAA)中的关键环节——最佳装配位姿拟合问题分为两步:第1步利用基于奇异值分解的解析方法将测量坐标系与装配现场的全局坐标系进行精确的空间配准,减小了坐标系对齐的误差,并以参考点拟合的偏差为优化目标,求解移动装配体当前位姿;第2步根据装配关键特性相关公差的重要程度,计算装配综合精度要求,并以最小综合偏差为优化目标求解移动装配体间的最佳装配位姿。随后给出了上述两个步骤的粒子群优化算法模型,将每步的待求解位姿作为一个拥有3个旋转自由度与3个平移自由度的粒子进行求解。最后对卫星舱段位姿最优装配问题进行仿真计算,结果证明了该优化算法在控制各项关键特性、提高综合装配质量等方面的有效性。
To control the key assembly characteristics and attain the best fit of posture with the aid of large-scale measure- ment, a multiple-objective optimization algorithm is proposed based on key assembly characteristics to posture best fit for large component assembly. This approach regards the posture best fit, which is a key activity of large component assembly in measurement-assisted assembly （MAA）, as a two-phase optimal problem. In the first phase, the measurement coordinate system and the global coordinate system are unified with minimum error based on singular value decomposition, and the cur- rent posture of the components being assembled is optimally solved in terms of the minimum variation of all the reference points. In the second phase, the synthetic error requirement of assembly is computed according to the importance of every key assembly characteristic to the whole product, and the best posture of the movable components are optimally determined by minimizing the synthetic error. Then the optimal model and the process procedures for the two phases based on particle swarm optimization are proposed. In this model, each posture to be calculated is modeled as a six dimensional particle （with three movement and three rotation parameters）. Finally, an assembly example about the joining of two cabin sections of a satellite mainframe structure is performed to verify the effectiveness of the proposed approach and process algorithm. The experiment result shows that the algorithm is robust and effective to control every key assembly characteristic and improve the assembly quality.
出处
《航空学报》
EI
CAS
CSCD
北大核心
2012年第9期1726-1736,共11页
Acta Aeronautica et Astronautica Sinica
基金
国家自然科学基金(51175026)~~
关键词
关键特性
大尺寸测量
测量辅助装配
位姿
最优拟合
粒子群算法
key characteristics
large-scale measurement
measurement-assisted assembly~ posture
best fit
particleswarm optimization
