Imperialist Competitive Algorithm with Effective Assimilation Strategy: A Comparative Study on Numerical Benchmark Functions

ABSTRACT

In recent years, the nature-inspired optimization algorithms known as intelligent optimization methods have been applied successfully for solving the different problems, along with the well-known mathematical methods. One of the new evolutionary algorithms presented lately is the imperialist competitive algorithm (ICA). This algorithm is based on the behaviour of imperialists in their attempt to conquer the colonies. In this paper, the original ICA has been extended and a new version of ICA has been proposed entitled “MICA”. The proposed MICA uses an efficient assimilation strategy to enhance the global exploration ability and to preserve a premature convergence. This new assimilation scheme uses the most powerful imperialist’s information to update. To validate the efficiency of the proposed algorithm, MICA is tested on a set of 28 non-linear benchmark functions with various dimensions and complexities. The results demonstrate that the proposed strategy enables the modified ICA to have better or at least comparable outcomes in comparison with the original ICA and the other state-of-the-art approaches at handing different types of problems.

KEYWORDS:

• Assimilation strategy
• Evolutionary algorithms
• Imperialist competitive algorithm
• Modified imperialist competitive algorithm
• Time-varying coefficients

Additional information

Notes on contributors

Elnaz Davoodi

Elnaz Davoodi received the BSc degree from Azarbaijan Shahid Madani University (ASMU), Tabriz, Iran, in 2008 and the MSc degree from the University of Tabriz, Tabriz, Iran, in 2012, both in electrical engineering. She is currently pursuing the PhD degree in electrical engineering at the University of Tabriz. Her research interests include the optimization, modelling, and power systems operation. Email: [email protected]

Ebrahim Babaei

Ebrahim Babaei received the PhD degree in electrical engineering from University of Tabriz, in 2007. In 2007, he joined the Faculty of Electrical and Computer Engineering, University of Tabriz. He has been Professor since 2015. He is the author and co-author of more than 430 journal and conference papers. He also holds 23 patents in the area of power electronics. His current research interests include the analysis, modelling, design, and control of power electronic converters and their applications, renewable energy sources, and FACTS devices. Babaei has been the Editor-in-Chief of the Journal of Electrical Engineering of the University of Tabriz, since 2013. He is also currently an Associate Editor of the IEEE Transactions on Industrial Electronics and IEEE Transactions on Power Electronics. He has been the Corresponding Guest Editor for different special issues in the IEEE Transactions on Industrial Electronics. In addition, Babaei has been the Track Chair, organizer of different special sessions and Technical Committee member in most important international conferences organized in the field of Power Electronics. Several times, he was the recipient of the Best Researcher Award from the University of Tabriz. Babaei has been included in the Top One Percent of the World's Scientists and Academics according to Thomson Reuters’ list in 2015, 2016 and 2017. From 1 October until 30 December 2016, he has been a Visiting Professor at the University of L’Aquila, Italy.

Behnam Mohammadi-ivatloo

Behnam Mohammadi-ivatloo received his BSc degree in electrical engineering from University of Tabriz, Tabriz, Iran in 2006, the MSc and PhD degrees from the Sharif University of Technology, Tehran, Iran in 2008 and 2012, respectively. He was a research associate at the University of Calgary from June 2011 to March 2012. He is currently an Associate Professor in the Faculty of Electrical and Computer Engineering, University of Tabriz, Tabriz, Iran. His main area of research is operation of modern power systems. Email: [email protected]