点击次数：

论文类型：期刊论文

发表时间：2016-10-01

发表刊物：ADVANCES IN SPACE RESEARCH

收录刊物：SCIE、EI、Scopus

卷号：58

期号：7

页面范围：1331-1347

ISSN号：0273-1177

关键字：Surrogate model; Adaptive sampling; Multi-objective optimization; Libration point; Optimal control; Invariant manifold

摘要：An adaptive surrogate model-based multi-objective optimization strategy that combines the benefits of invariant manifolds and low thrust control toward developing a low-computational-cost transfer trajectory between libration orbits around the L-1 and L-2 libration points in the Sun-Earth system has been proposed in this paper. A new structure for a multi-objective transfer trajectory optimization model that divides the transfer trajectory into several segments and gives the dominations for invariant manifolds and low-thrust control in different segments has been established. To reduce the computational cost of multi-objective transfer trajectory optimization, a mixed sampling strategy-based adaptive surrogate model has been proposed. Numerical simulations show that the results obtained from the adaptive surrogate-based multi-objective optimization are in agreement with the results obtained using direct multi-objective optimization methods, and the computational workload of the adaptive surrogate-based multi-objective optimization is only approximately 10% of that of direct multi-objective optimization. Furthermore, the generating efficiency of the Pareto points of the adaptive surrogate-based multi-objective optimization is approximately 8 times that of the direct multi-objective optimization. Therefore, the proposed adaptive surrogate-based multi-objective optimization provides obvious advantages over direct multi-objective optimization methods. (C) 2016 COSPAR. Published by Elsevier Ltd. All rights reserved.