UMP Institutional Repository

Novel Adaptive Spiral Dynamics Algorithms for Global Optimization

Ahmad Nor Kasruddin, Nasir and Tokhi, M. O. and Nur Maniha, Abd Ghani and Raja Mohd Taufika, Raja Ismail (2012) Novel Adaptive Spiral Dynamics Algorithms for Global Optimization. In: Proceedings of 11th IEEE Conference on Cybernetic Intelligence Systems 2012, 23 – 24 August 2012 , Limerick, Ireland. pp. 99-104..

[img]
Preview
PDF (fkee-2012-kasruddin-adaptive spiral dynamic)
fkee-2012-kasruddin-adaptive_spiral_dynamic.pdf - Published Version

Download (1MB)

Abstract

This paper presents adaptive versions of spiral dynamics algorithm (SDA) referred to as adaptive SDA (ASDA). SDA is known as fast computing algorithm due to its simplicity in the structure and it has stable convergence response when approaching the optimum point in the search space. However, the performance of SDA is still poor due to incorporation of single radius value during the whole search process. In ASDA, the spiral radius is made dynamic by employing novel mathematical equations and incorporating non-mathematical fuzzy logic strategy establishing the relationship between fitness value and spiral radius. This results in better performance in terms of convergence speed, accuracy, and total computing time while retaining the simple structure of SDA. Several uni-modal and multi-modal benchmark functions are employed to test the algorithm in finding the global optimum point. The results show that ASDA outperforms SDA in all test functions considered.

Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: Adaptive spiral dynamics; Optimization algorithm; Nature inspired; Fuzzy logic
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering
Faculty/Division: Faculty of Electrical & Electronic Engineering
Depositing User: Mr. Ahmad Nor Kasruddin Nasir
Date Deposited: 23 Sep 2013 02:43
Last Modified: 14 Mar 2018 07:44
URI: http://umpir.ump.edu.my/id/eprint/4224
Download Statistic: View Download Statistics

Actions (login required)

View Item View Item