Please use this identifier to cite or link to this item:
|Title:||DIFFERENTIAL EVOLUTION-BASED METHODS FOR NUMERICAL OPTIMIZATION||Authors:||QIU XIN||Keywords:||Evolutionary Algorithm,Differential Evolution,Multi-objective Optimization,Minimax Optimization,Numerical Optimization,Robust Optimization||Issue Date:||11-Aug-2016||Citation:||QIU XIN (2016-08-11). DIFFERENTIAL EVOLUTION-BASED METHODS FOR NUMERICAL OPTIMIZATION. ScholarBank@NUS Repository.||Abstract:||This thesis focuses on designing new Differential Evolution (DE) variants to overcome the limitations of existing approaches in solving single-objective, multi-objective and minimax optimization problems. First, a multiple exponential recombination strategy is proposed to fill the research gap in DE crossover operator development. The new strategy is able to inherit all the main advantages of existing crossover operators while possessing a stronger ability in handling dependent variables. Next, a novel multi-objective DE algorithm is designed for circumventing the convergence slowdown, parametric sensitivity and lack of flexibility issues while extending DE to multi-objective optimization. Last, the minimax optimization problems in robust design are investigated. A new minimax DE algorithm is developed to efficiently address the fundamental issues in current minimax optimization area. Besides the statistical superiority over other state-of-the-art approaches, the new algorithm is also successfully applied to solve two open problems in robust control.||URI:||http://scholarbank.nus.edu.sg/handle/10635/134464|
|Appears in Collections:||Ph.D Theses (Open)|
Show full item record
Files in This Item:
|QiuX.pdf||8.71 MB||Adobe PDF|
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.