首页> 外文期刊>Chinese science bulletin >Constructing orthogonal fractional factorial designs with the same confounding structure
【24h】

Constructing orthogonal fractional factorial designs with the same confounding structure

机译:使用相同的混杂结构构造正交分数阶因子设计

获取原文
获取原文并翻译 | 示例
       

摘要

ORTHOGONAL arrays (OAs) have important applications to many fields. A lot of OAs and their construction methods have been introduced in the experimental design literature (for example, Box, Hunter and Dey ). These OAs are used for an experimenter to choose in practice. But it often happens that any existing OA is not suitable for a given experiment, i.e. any existing OA contains some infeasible factor-level combinations for the experiment to be executed. In this case, it is necessary to consider a flexible method to generate a suitable OA. Cheng and Li and Wang and Zhang give a necessary and sufficient condition, of existence of OA debarring some infeasible combinations and some construction methods. But these methods need large computation and it is difficult to carry out in practice. It fact, there are many OAs with the same defining relations, but in the existing literature usually only one is given and not debarring the specific infeasible factorlevel combinations. It is possible to find a suitable one from all the OAs with the same defining relations. With this consideration, we can reduce the computations to pick up an OA debarring the infeasible factor-level combinations. This note will give a method of constructing such OAs from all the OAs with the same confounding structure.
机译:正交数组(OA)在许多领域都有重要的应用。实验设计文献(例如Box,Hunter和Dey)中引入了许多OA及其构造方法。这些OA用于实验人员在实践中进行选择。但是,经常会发生这样的情况,即任何现有的OA都不适合给定的实验,即任何现有的OA都包含一些无法执行的因子级组合以执行该实验。在这种情况下,必须考虑一种灵活的方法来生成合适的OA。 Cheng和Li以及Wang和Zhang给出了OA的存在的必要和充分条件,禁止了一些不可行的组合和某些构造方法。但是这些方法需要大量的计算,并且在实践中很难执行。实际上,有许多具有相同定义关系的OA,但是在现有文献中通常只给出一个,并且不排除特定的不可行因素级别组合。可以从具有相同定义关系的所有OA中找到合适的对象。考虑到这一点,我们可以减少计算量,以取消不可行的因子水平组合。本说明将提供一种从所有具有相同混淆结构的OA构造此类OA的方法。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号