An intelligent method for early design and decision of space payload

This paper is concerned with the design and decision method of space payload in the early design phase. The objective is to develop a generalized intelligent framework to assist designers in designing space payload. A multi-level method is introduced, which is based on the combination of case-based reasoning (CBR), multicriteria rough sets theory (MRST) and kernel-based self-organizing maps (KSOM). The first is CBR module, which aims to help designers to obtain some similar cases of the new situation. The second is MRST module, which can extract the decision rules of configuration selection. The third is KSOM module, which can reduce the variable range and design space of the space payload. The results demonstrate the effectiveness of the proposed design method.