2008
DOI: 10.1002/aic.11490
|View full text |Cite
|
Sign up to set email alerts
|

Reliable computation of equilibrium cascades with affine arithmetic

Abstract: Computing the steady state of multistage counter-current processes like distillation, extraction, or absorption is the equivalent to finding solutions for large scale non-linear equation systems. The conventional solution techniques are fast and efficient if a good estimation is available but are prone to fail, and do not provide information about the reason for the failure. This is the main motive to apply reliable methods in solving them. Topical heading: Separations

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
10
0

Year Published

2009
2009
2016
2016

Publication Types

Select...
5

Relationship

2
3

Authors

Journals

citations
Cited by 6 publications
(10 citation statements)
references
References 50 publications
0
10
0
Order By: Relevance
“…The mixed AA/IA was used only at the critical parts (where otherwise division by zero or calling the logarithm function with negative argument would have occurred) in the previous work41 of the authors due to implementation design flaws. Based on the conclusions of the previous work, the affine class has been redesigned and implemented in C++.…”
Section: Procedures For Locating All Solutionsmentioning
confidence: 99%
See 3 more Smart Citations
“…The mixed AA/IA was used only at the critical parts (where otherwise division by zero or calling the logarithm function with negative argument would have occurred) in the previous work41 of the authors due to implementation design flaws. Based on the conclusions of the previous work, the affine class has been redesigned and implemented in C++.…”
Section: Procedures For Locating All Solutionsmentioning
confidence: 99%
“…It follows that only the first LP subproblem has to be solved from scratch; all other subproblems should use the optimal solution of the preceding subproblem as a primal feasible basis and run only Phase II of the primal simplex algorithm, thus hopefully reducing the computational efforts. The naïve sequence41 to process the x j variables would be min x 1 , max x 1 , min x 2 , max x 2 , … etc but the subproblems min x 1 and max x 1 are likely to produce completely different solutions. As a consequence, this is expected to result in a lot of simplex iterations when using the optimal solution of subproblem min x 1 as the initial primal feasible basic solution when solving the subproblem max x 1 .…”
Section: Procedures For Locating All Solutionsmentioning
confidence: 99%
See 2 more Smart Citations
“…One of the characteristic components of the proposed algorithm is the linearization technique. Numerical evidence published in the literature30–42 seem to indicate superiority of the linear enclosure compared with the traditional one such as the interval Newton methods [see Ref 43…”
Section: Methodsmentioning
confidence: 99%