Key laboratory of Advanced Design and Intelligent Computing, Ministry of Education, School of Software Engineering, Dalian University, Dalian, China.
School of Engineering, Computer and Mathematical Sciences, Auckland University of Technology, Auckland 1010, New Zealand.
Math Biosci Eng. 2022 Sep 26;19(12):14142-14172. doi: 10.3934/mbe.2022659.
Metaheuristic algorithms have the drawback that local optimal solutions are prone to precocious convergence. In order to overcome the disadvantages of the whale optimization algorithm, we propose an improved selective opposition whale optimization algorithm (ISOWOA) in this paper. Firstly, the enhanced quasi-opposition learning (EQOBL) is applied to selectively update the position of the predator, calculate the fitness of the population before and after, and retain optimal individuals as the food source position; Secondly, an improved time-varying update strategy for inertia weight predator position is proposed, and the position update of the food source is completed by this strategy. The performance of the algorithm is analyzed by 23 benchmark functions of CEC 2005 and 15 benchmark functions of CEC 2015 in various dimensions. The superior results are further shown by Wilcoxon's rank sum test and Friedman's nonparametric rank test. Finally, its applicability is demonstrated through applications to the field of biological computing. In this paper, our aim is to achieve access to DNA files and designs high-quantity DNA code sets by ISOWOA. The experimental results show that the lower bounds of the multi-constraint storage coding sets implemented in this paper equals or surpasses that of previous optimal constructions. The data show that the amount of the DNA storage cods filtered by ISOWOA increased 2-18%, which demonstrates the algorithm's reliability in practical optimization tasks.
元启发式算法的缺点是容易出现局部最优解过早收敛。为了克服鲸鱼优化算法的缺点，本文提出了一种改进的选择性对立鲸鱼优化算法（ISOWOA）。首先，采用增强型拟对立学习（EQOBL）选择性地更新捕食者的位置，计算种群前后的适应度，保留最优个体作为食物源位置；其次，提出了一种改进的时变惯性权重捕食者位置更新策略，通过该策略完成食物源的位置更新。通过 CEC 2005 的 23 个基准函数和 CEC 2015 的 15 个基准函数在不同维度上对算法性能进行了分析。通过 Wilcoxon 秩和检验和 Friedman 非参数秩检验进一步显示了优越的结果。最后，通过在生物计算领域的应用证明了其适用性。在本文中，我们的目的是通过 ISOWOA 访问 DNA 文件和设计高质量的 DNA 代码集。实验结果表明，本文实现的多约束存储编码集的下界等于或超过了以前的最优构造。数据表明，ISOWOA 过滤的 DNA 存储码数量增加了 2-18%，这证明了该算法在实际优化任务中的可靠性。
