期刊文献+

试验设计类型之可以考察部分交互作用的多因素设计:正交设计与均匀设计 被引量:8

Types of the multifactor experimental designs with partial interactions:orthogonal and uniform design
下载PDF
导出
摘要 本文旨在介绍两种多因素试验设计方法,即正交设计与均匀设计。由于析因设计中的试验点数目过多,在既要较少试验点数目,又要能比较准确地揭示多因素对评价指标影响规律的要求之下,宜选择正交设计或均匀设计。正交设计具有"试验点均匀分散、整齐可比且可以事先安排少数交互作用项"的特点,因此,结果的可信度较高,只需采用方差分析处理定量资料;均匀设计具有"试验点极其均匀分散、正交性在一定程度上受到破坏,不能事先安排交互作用项"的特点,因此,允许很少数目的试验点,但需采用多重回归分析处理资料且分析结果不唯一。 The paper aims to introduce two multifactor experimental designs orthogonal design and uniform design. Excessive experimental points are required in factorial design, that may be its limitations. Orthogonal design and uniform design are frequently applied, because less experimental points are required in these designs. Additionally, they may demonstrate the effects of experimental factors on evaluation index accurately. Orthogonal design is characterized by uniform dispersion and regularly - comparable experimental points. Additionally, it may arrange partial interactions in advance. Consequently, analysis of variance may be applied to analyze data from Orthogonal design and the results are reliable. Experimental points in uniform design are extremely uniformly distributed at the cost of reduce orthogonality. Uniform design may not arrange interactions in advance. Consequently, multiple regression analysis may be utilized to analyze data from uniform design, and the results may be various.
出处 《四川精神卫生》 2017年第3期205-210,共6页 Sichuan Mental Health
基金 国家高技术研究发展计划课题资助(2015AA020102)
关键词 析因设计 分式析因设计 正交设计 均匀设计 试验点 Factorial design Fractional factorial design Orthogonal design Uniform design Experimental point
  • 相关文献

同被引文献109

引证文献8

二级引证文献33

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部