NEW TECHNIQUE FOR SOLVING MULTIVARIATE GLOBAL OPTIMIZATION
Journal of Numerical Analysis and Approximation Theory, cilt.52, sa.1, ss.3-16, 2023 (Scopus)
- Yayın Türü: Makale / Tam Makale
- Cilt numarası: 52 Sayı: 1
- Basım Tarihi: 2023
- Doi Numarası: 10.33993/jnaat521-1287
- Dergi Adı: Journal of Numerical Analysis and Approximation Theory
- Derginin Tarandığı İndeksler: Scopus
- Sayfa Sayıları: ss.3-16
- Anahtar Kelimeler: branch and bound method, convex underesti-mation, explicit solution, global optimization, piecewise quadratic
- Kırklareli Üniversitesi Adresli: Evet
Özet
In this paper, we propose an algorithm based on the Branch and Bound method to underestimate the objective function and reductive transformation which transformed all multivariable functions into univariate functions. We also propose and demonstrate several quadratic lower bound functions, which are better/preferable to the ones mentioned in the literature. In this regard, our experimental results are more effective when we face different nonconvex functions.