Adaptive Mechanism For Enhanced Performance Of Shark Smell Optimization

Abstract

Various meta-heuristic approaches have been developed to find the optimal solution to optimization problems. However, different approach takes different amount of time and efficiency to achieve the optimal solution. Determination of high performance and lower computing time with simple algorithm is therefore continuously established. Shark Smell Optimization (SSO) algorithm has been proven to have high efficiency in many optimization applications. However, like the other swarm intelligence, SSO algorithm also has possibility to get trapped in local maxima or premature convergence. Thus, a new adaptive shark smell optimization (ASSO) is proposed to improve the convergence efficiency of standard SSO algorithm. An overview and performance comparison of six well-known meta-heuristic optimization algorithm is also presented in this paper. In order to verify the effectiveness of this newly developed method, the algorithm was tested on common benchmark functions used in the literature. Numerical results indicate that the ASSO algorithm strategy outperforms the basic SSO algorithm, Genertic Algorithm (GA), Particle Swarm Intelligence (PSO), Firefly Algorithm (FA), Artificial Bee Colony (ABC) and Teaching Learning Based Optimization (TBLO) in term of reaching for global solution.