基于灰狼算法的改进研究

针对灰狼算法具有易陷于局部最优并且收敛速度不理想的缺点,本文提出基于改进收敛因子策略和引入动态权重策略以及两种策略混合改进的灰狼优化算法,并且用于求解函数优化问题。提出的一种非线性收敛因子公式,能够动态的调整算法的全局搜索能力,引入的动态权重使算法在收敛过程中能够加快算法的收敛速度。通过15个基准测试函数进行验证改进后的算法的全局搜索能力、局部搜索能力与收敛速度,实验结果表明:改进后的算法无论在搜索能力上还是收敛速度上,都强于标准灰狼算法。

Study on Improvement of based on the gray wolf algorithm

According to gray wolf algorithm is easily trapped in local optimum and the disadvantage of the convergence rate is not ideal, is presented in this paper based on the improved convergence factor strategy and dynamic weighting strategies and two mixed strategy improved the wolf optimization algorithm was introduced and used to solve the function optimization problem. A nonlinear convergent factor formula is proposed, which can dynamically adjust the global searching ability of the algorithm, and the dynamic weight is introduced to accelerate the convergence speed of the algorithm. Through 15 benchmark test functions to verify the improved algorithm of the global search ability and local search ability and convergence speed. The experimental results show that: the improved algorithm in terms of search ability or convergence rate, strong in the standard Wolf algorithm.