%0 Journal Article
%T 基于蜻蜓算法的改进研究<br>Study on Improvement of Dragonfly Algorithm
%A 胡小平
%A 周非无
%J Computer Science and Application
%P 1377-1386
%@ 2161-881X
%D 2019
%I Hans Publishing
%R 10.12677/CSA.2019.97155
%X <div style=\"text-align:justify;\">
	<span style=\"font-size:14px;\">针对标准蜻蜓算法中存在的收敛速度慢，易于局部解的缺点，提出一种改进的蜻蜓算法(IDA)。该算法提出两种非线性函数，分别动态调节列队权重和聚集权重的收敛因子，提高算法平衡全局搜索和局部开发的能力；灰狼机制有较强的局部开发能力和收敛速度，融合灰狼机制以提高蜻蜓算法的收敛精度和速度；算法迭代后期种群多样性下降，引入末位淘汰策略来提高种群的多样性，使算法跳出局部解。通过6个复杂的测试函数对改进算法进行仿真，并和其他三个算法进行对比。结果表明，IDA算法的收敛精度、收敛速度和稳定性都优于其他三个算法。<br />
An improved dragonfly algorithm (IDA) was proposed to overcome the disadvantages of the standard dragonfly algorithm, such as slow convergence rate and easy to be trapped in local solutions. In order to improve the ability of balancing exploration and exploitation, IDA algorithm proposes two kinds of nonlinear function that can dynamically adjust the convergence factors of the alignment weight and cohesion weight. Grey Wolf mechanism has good performance in exploitation and rate of convergence. In order to improve the convergence accuracy and speed of the dragonfly algorithm, the grey Wolf mechanism was incorporated into the dragonfly algorithm. In the late iteration of the algorithm, the diversity of the population decreases, which makes the algorithm easy to fall into the local solution. The lowliest place elimination series is introduced to improve the diversity of the population and make the algorithm jump out of the local solution. The improved algorithm is simulated with six complex functions and compared with the other three algorithms. The results show that the convergence accuracy, convergence speed and stability of IDA algorithm are better than the other three algorithms.</span> 
</div>
%K 蜻蜓算法，非线性函数，灰狼机制，末位淘汰<br>Dragonfly Algorithm
%K Nonlinear Function
%K Grey Wolf Mechanism
%K Lowliest Place Elimination Series
%U http://www.hanspub.org/journal/PaperInformation.aspx?PaperID=31406