Search

Space station logistics strategy optimisation is a complex engineering problem with multiple objectives. Finding a decision-maker-preferred compromise solution becomes more significant when solving such a problem. However, the designer-preferred solution is not easy to determine using the traditional method. Thus, a hybrid approach that combines the multi-objective evolutionary algorithm, physical programming, and differential evolution (DE) algorithm is proposed to deal with the optimisation and decision-making of space station logistics strategies. A multi-objective evolutionary algorithm is used to acquire a Pareto frontier and help determine the range parameters of the physical programming. Physical programming is employed to convert the four-objective problem into a single-objective problem, and a DE algorithm is applied to solve the resulting physical programming-based optimisation problem. Five kinds of objective preference are simulated and compared. The simulation results indicate that the proposed approach can produce good compromise solutions corresponding to different decision-makers’ preferences. [ABSTRACT FROM AUTHOR]

Rendezvous and docking (RVD) is a key technology for performing complicated space missions. After an RVD process, several activities are executed to consolidate two docked spacecraft into a spacecraft complex, and this task phase is referred to as a spacecraft consolidation mission. It can save the mission time to execute these activities in parallel, but a high degree of parallelism could result in a disordered execution profile and many violations of precedence constraints. To solve this contradiction, a hybrid multi-objective optimisation approach is proposed. The precedence requirements within each activity are satisfied using an encoding and scheduling process, while the precedence requirements between different activities are treated by adding release time variables. A compact-execution index is designed to express the preference of an orderly and compact execution profile. Furthermore, a multi-objective hybrid-encoding genetic algorithm is employed to find optimal solutions. Finally, the proposed approach is demonstrated for a numerical example. The results show that optimal solutions satisfying precedence requirements both within each activity and between different activities are successfully obtained, and the trade-off between saving mission time and obtaining an orderly and compact execution profile can be effectively made. The performance of the proposed method is validated by comparison with two other multi-objective genetic algorithms. [ABSTRACT FROM PUBLISHER]