基于微分进化的行星际小推力轨道全局优化方法

Global Optimization of Interplanetary Low-Thrust Trajectory Based on Differential Evolution

针对行星际小推力轨道优化问题,提出一种基于改进微分进化的全局优化算法.通过引入试验个体重生成和约束判断选择策略,克服传统微分进化算法中寻优参数和轨道约束违反边界的缺陷.为提高微分进化算法后期收敛效率,提出了基于最优个体信息的变异操作和局部搜索辅助策略.以地球-水星的小推力燃料最省转移为例对所提算法进行了验证.数值计算结果表明:改进的微分进化算法能够快速有效地寻找到全局最优轨道,并且与传统非线性规划和遗传算法相比,具有更高的可靠性和收敛性.

A novel global optimization approach for interplanetary low-thrust trajectory based on modified differential evolution is proposed. First, the trial vector regeneration strategy and constraint selection criteria are introduced into differential evolution process, which can keep the optimization variables and trajectory constraints from boundary violation. Then, in order to improve the convergence efficiency of differential evolution close to the global optimum, a new mutation operation with best individual information and an assistant local optimization strategy are proposed. Taking optimal-fuel low-thrust transfer from earth to mercury as an example, the modified differential evolution algorithm （MDEA） is validated. The simulation results demonstrate that the proposed algorithm is effective to find the global optimal transfer trajectory, and have higher reliability and better convergence ability comparing to sequential quadratic programming and genetic algorithm.