An Improved Differential Evolution Algorithm and its Application in Reaction Kinetic Parameters Estimation
Dan Xu, Shaojun Li, Feng Qian · 2007
Differential evolution algorithm (DE) is a simple efficient optimizationtechnique, but it is easily trapped in the local optima.This paper presents an improved differential algorithm (IDE) based on Alopex (Algorithms of Pattern Extraction) where "noise" strategy according to the learning experience and memory selection are used.In order to seek for better result Alopex operator contracts searching area self-adaptively during iteration process.The performance of IDE is tested by several benchmark functions.Results show that the IDE algorithm overcomes the disadvantages of the original DE and possesses higher precision.Finally IDE is successfully applied to reaction kinetic parameters estimation.