应用物理规划方法对500米口径球面射电望远镜(FAST)精调Stewart平台进行了多目标优化设计。根据Stewart平台的设计准则,将Stewart平台的运动精度、重量和工作效率作为目标函数,构造了相应的以目标函数为变量的偏好函数,建立了基于物理规划的多目标优化模型,采用遗传算法对该优化问题进行求解。结果表明,物理规划避免了基于权重的多目标优化方法中权系数难以确定的问题,有效地给出了优化问题的Pareto解。
The physical programming method was applied to the multi-objective optimization of the Stewart platform used for the 500-meter aperture spherical radio telescope (FAST). Based on the design criteria for the Stewart platform, kinematic accuracy, weight and working efficiency of the Stewart platform were considered as objective functions, and their relevant preference functions were constructed using physical programming. Then a multi-objective optimization model for the Stewart platform was established and the optimization was solved by the genetic algorithm. The results show that the physical programming can avoid the problem of determining weight factors in weight-based multi-objective optimization approaches, and give a Pareto optimal solution of multi-objective optimization effectively.