|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|93195||2017||13 صفحه PDF||سفارش دهید||8373 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Swarm and Evolutionary Computation, Available online 6 December 2017
An Enhanced Imperialist Competitive Algorithm (EICA) is proposed aimed at enabling the ICA algorithm to escape from local optima faster, thus enabling faster convergence of the basic ICA algorithm. To this end, the imperialistic competition phase of the algorithm is enhanced by giving added value to a slightly unfeasible solution, based on its distance from the relative imperialist. The performance of the proposed EICA algorithm is investigated through design optimizations of four benchmark side-sway frames. Results indicate that, in terms of both the design quality and the solution speed, EICA compares significantly favourable with a number of other meta-heuristic optimizers, including the basic ICA.