Environmental Engineering Reference
In-Depth Information
Figure 2. Example of sub-region of full simplex containing range of feasible mixtures.
x 3 2 = x 3 .(1- x 1 - x 2 )
Since feasible brick mixtures do not exist over the entire region as shown
in Figure 1, a sub region of the full simplex containing the range of feasible
mixtures must be defined by constraining the component proportions. An
example of a possible sub region for the three components is shown in Figure
2 and is defined by the following mass fractions (where x 1 = jarosite waste, x 2
= clay, x 3 = CCRs);
0.15  x 1  0.25
0.10  x 2  0.20
0.60  x 3  0.70
1.4. Model Fitting and Validation
To verify the adequacy of the chosen model quantitatively and
graphically, the polynomial models described in the classic mixture approach
are fit to data using analysis of variance (ANOVA) and least squares
techniques (Box, 1978). Many statistical software packages have the capability
to perform these analyses and data fitting. From the ANOVA significance of
the treatment effect can also be obtained. Although the polynomial models are
slightly different for the classical mixture approach many of the steps involved
in model selection and fitting are the same. To select the suitable model
(linear, quadratic, etc.) the primary aspect to be performed is establishing
Search WWH ::




Custom Search