A center-cut algorithm for solving convex mixed-integer nonlinear programming problems

A4 Konferenspublikationer

Interna författare/redaktörer

Publikationens författare: Jan Kronqvist, Andreas Lundell, Tapio Westerlund
Redaktörer: Antonio Espuña, Moisès Graells, Luis Puigjaner
Publiceringsår: 2017
Förläggare: Elsevier
Moderpublikationens namn: 27 European Symposium on Computer Aided Process Engineering
Seriens namn: Computer Aided Chemical Engineering
Volym: 40
Artikelns första sida, sidnummer: 2131
Artikelns sista sida, sidnummer: 2136
ISBN: 978-0-444-63965-3
eISBN: 9780444639707
ISSN: 1570-7946


    In this paper, we present a new algorithm for solving convex mixed-integer nonlinear programming problems. Similarly to other linearization-based methods, the algorithm generates a polyhedral approximation of the feasible region. The main idea behind the algorithm is to use a different approach for obtaining trial solutions. Here trial solutions are chosen as a center of the polyhedral approximation. By choosing the trial solutions as such, the algorithm is more likely to obtain feasible solutions within only a few iterations, compared to the approach of choosing trial solutions as the minimizer of a linear approximation of the problem. The algorithm can be used both as a technique for finding the optimal solution or as a technique for quickly finding a feasible solution to a given problem. The algorithm has been applied to some challenging test problems, and for these the algorithm is able to find a feasible solution within only a few iterations.

    Senast uppdaterad 2020-02-10 vid 01:28