A sequential quadratically constrained quadratic programming technique for a multi-objective optimization problem

In this article a line search algorithm is proposed for solving constrained multi-objective optimization problems. At every iteration of the proposed method, a subproblem is formulated using quadratic approximation of all functions. A feasible descent direction is obtained as a solution of this subp...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Engineering optimization Ročník 51; číslo 1; s. 22 - 41
Hlavní autori: Ansary, Md. Abu Talhamainuddin, Panda, Geetanjali
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Abingdon Taylor & Francis 02.01.2019
Taylor & Francis Ltd
Predmet:
Algorithms
Constraints
efficient solution
Multi-objective optimization
Multiple objective analysis
Optimization
Penalty function
purity metric
Quadratic programming
Search algorithms
spread metric
ISSN:0305-215X, 1029-0273
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:In this article a line search algorithm is proposed for solving constrained multi-objective optimization problems. At every iteration of the proposed method, a subproblem is formulated using quadratic approximation of all functions. A feasible descent direction is obtained as a solution of this subproblem. This scheme takes care some ideas of the sequential quadratically constrained quadratic programming technique for single objective optimization problems. A non-differentiable penalty function is used to restrict constraint violations at every iterating point. Convergence of the scheme is justified under the Slater constraint qualification along with some reasonable assumptions. The proposed algorithm is verified and compared with existing methods with a set of test problems. It is observed that this algorithm provides better results in most of the test problems.
Bibliografia:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0305-215X
1029-0273
DOI:10.1080/0305215X.2018.1437154