|
Mixture Experiments in industrial and agricultural production , scientific research and management applications in a very wide range . Mixing test is a special type of test , in which the response of each mixture component is only a function of the proportion of , and not a function of the total mixture . The q- component mixing system , the test indicator is the proportion of the components x1, x2, · · ·, xq a function of the component ratio of each row to satisfy the constraint ( ? ) , Where ( ? ) Geometric interpretation ( ? ) formal simple dimensional Euclidean space . Satisfy this constraint conditions on the mixing system , the ordinary polynomial model can not directly fit this mixing problem . Many models have been proposed mixing and mixing design methods. Scheffe ' in 1958 and in 1963 he proposed mixing plaid design and center of gravity design , these designs is the border design, that in addition to all of the center of gravity , other design points are simple type of boundary ( vertices , edges, faces ) on . To make the test points as much as possible in simple interior , Cornell in 1975 proposed shaft design, the Cox method applied to the mixing multiple linear polynomial model , defines all the test points in q -dimensional simplex connecting the center of gravity and the design of the shaft of the vertices . Mixing shaft design of the data can be used as a tool to describe the effects of components , Sq-1 along the i axis , all the other j (j = i) mixing components of a change at the same rate . Furthermore, in some areas of mixture experiments , particularly in the production of chemical industry , often a number of important components for people to choose , and the use of shaft design, these components can be screened to select the most important element to improve the economic benefits. Q component of interest in the area of simple regular Sq-1 , for the second-order plus q component mixture model , the authors use symmetry axis design , matrix algebra , numerical methods , A? Optimal design theory , which can be discussed and plus shaft design class mixture model on a? optimal axial designs. This paper is divided into three chapters. The first chapter introduces the mixing problem , mixture design design methods commonly used . The second chapter introduces five , six second additive component mixing model A? Optimality . Chapter III using computer-aided method to second shaft design for additive mixture model on the class A? Optimalities promotion to general .
|