Abstract: Addressing the challenges in complex system architecture design,such as vast design spaces,numerous alternatives,and intricate evaluation metrics that complicate trade-offs among architectural options,a novel methodology for solving and balancing complex system architecture designs was proposed.The approach integrated Constraint Satisfaction Problem (CSP) and Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) based weight sensitivity analysis to generate satisfactory architectural solutions that met the needs of multiple stakeholders.Initially,the decision variables of the system architecture were identified to construct a morphological matrix,which helped in defining the metrics for architectural decision-making.Subsequently,a CSP model was developed based on constraint information to reduce the architectural space,and a multi-objective optimization was employed to determine the Pareto frontier.Following this,the solutions were ranked and filtered using weight sensitivity analysis combined with the TOPSIS method,leading to the selection of the most satisfactory architectural solution.Finally,the effectiveness of the proposed method was demonstrated through a case study of the design of the primary and secondary separation system of the launch vehicle.

Key words: complex system, architectural decision, constraint satisfaction problem, multi-objective optimization, trade-off, primary and secondary separation system

摘要： 针对复杂系统架构设计面临的设计空间庞大、备选项多、评估指标复杂导致架构方案难以权衡的问题,提出一种支持复杂系统架构设计的求解与权衡方法,通过约束满足问题(CSP)和基于理想点法(TOPSIS)的权重敏感性分析方法,生成满足多方利益相关者的满意架构方案。通过识别系统架构的决策项以构建形态矩阵,确定架构决策的衡量指标;基于约束信息构建CSP模型以约简架构空间,并通过多目标优化确定Pareto前沿;结合权重敏感性分析和TOPSIS方法对方案进行排序筛选,从而选出满意的架构方案。最后,通过运载火箭一二级分离系统设计案例验证了该方法的有效性。

关键词: 复杂系统, 架构决策, 约束满足问题, 多目标优化, 权衡, 一二级分离系统