Computer Engineering and Applications ›› 2007, Vol. 43 ›› Issue (31): 9-12.

• 博士论坛 • Previous Articles     Next Articles

Multi-objective compatible iterative control algorithm for nonconvex objective problem

HU Qing-song,XU Li-hong   

  1. School of Electronics and Information,Tongji University,Shanghai 200092,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-11-01 Published:2007-11-01
  • Contact: HU Qing-song

目标空间非凸的多目标相容迭代控制算法

胡庆松,徐立鸿   

  1. 同济大学 电子与信息工程学院,上海 200092
  • 通讯作者: 胡庆松

Abstract: Since it is difficult for the existing control algorithm to deal with the nonconvex objective space control problem,combining the latest research result of the multi-objective genetic optimization algorithm,a new multi-objective compatible algorithm is proposed.Considering the computing speed requirement in the control process,iterative algorithm is put forward.During the control process,by selection function,the controllability of nonconvex objective is guaranteed.The simulation example illuminates that compared with the previous methods,the new algorithm is effective in the nonconvex objective control.

摘要: 针对现有控制算法难以解决多目标空间为非凸的控制问题的现状,结合多目标遗传进化理论的最新成果,提出了多目标相容控制算法。考虑到控制过程中对计算速度的要求,利用控制问题状态具有连续性的特点,提出了迭代算法。在算法进行过程中,通过选择函数进行目标的选择,保证了非凸目标可控性。仿真实例说明了与原有控制算法相比,新算法在非凸目标可控性方面的有效性。