The SOS algorithm is an effective meta-heuristic algorithm, but it may lead to overexploration and has difficulties in balancing between exploration and exploitation capabilities. In this study, two extended versions of the SOS algorithm are proposed with different weight strategies. Experimental results show that the proposed algorithms provide promising results and outperform compared algorithms.
The symbiotic organisms search (SOS) algorithm is a current effective meta-heuristic algorithm, which is been applied to solve various types of optimization problems. However, the SOS can easily lead to overexploration in the parasitism phase, and it is difficult to balance between exploration and exploitation capabilities. In the present work, two extended versions of the SOS are proposed. Two different weight strategies (i.e., random-weight and adaptive-weight) are utilized to generate the weighted mutual vector, respectively. Meanwhile, the best organism is employed to produce the modified artificial parasite vector. The performance of the two improved algorithms is evaluated on 35 test functions. The results demonstrate that the proposed algorithms are able to provide very promising results. Furthermore, five real-world problems are solved by the two newly proposed methods. Experimental results demonstrate that the presented algorithms are more efficient than the compared algorithms. All the obtained results further indicate that the two proposed algorithms are competitive and provide better results when compared to a wide range of algorithms, including SOS and its five modified versions, as well as ten other meta-heuristic algorithms.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据