系统工程与电子技术 ›› 2026, Vol. 48 ›› Issue (4): 1243-1253.doi: 10.12305/j.issn.1001-506X.2026.04.14

• 系统工程 • 上一篇    

规范约束空间下多类型因子混合MP-CE试验设计方法

王子辰1(), 潘正强1,*, 王彦琳1, 金光1, 邹伟2   

  1. 1. 国防科技大学系统工程学院,湖南 长沙 410073
    2. 中南大学计算机学院,湖南 长沙 410083
  • 收稿日期:2023-11-06 修回日期:2024-08-23 接受日期:2026-03-18 出版日期:2024-12-06 发布日期:2024-12-06
  • 通讯作者: 潘正强 E-mail:wangzichen0128@126.com
  • 作者简介:王子辰(2000—),男,博士研究生,主要研究方向为系统试验与评估
    王彦琳(2001—),女,硕士研究生,主要研究方向为系统试验与评估
    金 光(1973—),男,研究员,博士,主要研究方向为寿命预测与健康管理、系统试验与评估、数据分析与建模
    邹 伟(1978—),男,讲师,博士,主要研究方向为面向领域大数据的分析、数据挖掘、机器学习
  • 基金资助:
    国家自然科学基金(72171231)资助课题

Multi-type factors mixed experimental design method of MP-CE in normalized constrained space

Zichen WANG1(), Zhengqiang PAN1,*, Yanlin WANG1, Guang JIN1, Wei ZOU2   

  1. 1. College of Systems Engineering,National University of Defense Technology,Changsha 410073,China
    2. School of Computer Science and Engineering,Central South University,Changsha 410083,China
  • Received:2023-11-06 Revised:2024-08-23 Accepted:2026-03-18 Online:2024-12-06 Published:2024-12-06
  • Contact: Zhengqiang PAN E-mail:wangzichen0128@126.com

摘要:

在装备试验设计中,多类型因子混合的情形非常普遍,同时各因子间还存在复杂的约束关系。针对如此复杂不规则的试验空间,提出一种规范约束空间下的最大投影坐标交换(maximum projection coordinate exchange, MP-CE)试验设计方法。首先,对典型的因子约束进行定义和规范化描述;其次,给出统一的试验空间描述方法,对解析约束和非解析约束等进行表示,并构建规范约束试验空间;然后,根据规范约束试验空间特点,提出改进的MP-CE算法,以有效解决复杂约束下多类型因子混合试验设计问题;最后,通过实例对比研究,从模型预测效果、空间填充性、最优性准则等多个评价指标,验证了所提方法的有效性和优势。

关键词: 试验设计, 多类型因子混合, 最大投影准则, 坐标交换, 约束空间

Abstract:

In the design of equipment experiments, the mixing of multi-type factors is very common, and there are complex constraint relations among the factors. For complex and irregular experiment space, a maximum projection coordinate exchange experimental design algorithm (MP-CE) in normalized constrained space is proposed. Firstly, typical factor constraints are defined and described in standardised terms. Then, a unified experiment space description method is presented to represent analytical constraints and non-analytical constraints, and construct normalized constrained experimental space. Then, the improved MP-CE algorithm is proposed according to the characteristics of normalized constrained experimental space to effectively addressing multi-factor mixed experimental design problems under complex constraints. Finally, a comparative study is carried out through an application example. The effectiveness and advantages of the proposed method are verified by calculating multiple evaluation indexes such as model prediction effect, space filling and optimality criterion.

Key words: design of experiment, multi-type factors mixing, maximum projection criterion, coordinate exchange, constraint space

中图分类号: