混合策略改进灰狼优化算法的函数优化

Improved Grey Wolf Optimizer for Function Optimization Using Hybrid Strategy

为了解决灰狼优化算法在函数优化过程中搜索精度不高的问题,提出了一种分群优化、高斯变异和随机扰动混合策略改进的灰狼优化算法。一方面,通过采用分群优化策略,加强算法局部搜索与全局搜索之间的信息交换;另一方面,采用高斯变异和随机扰动策略维持算法进化过程中的种群多样性,并利用贪婪思想更新种群。通过引入包含单峰、多峰和固定维度多峰的多个基准测试函数,仿真实验验证了所提改进灰狼算法有效性。在与其他几种先进优化算法的综合比较与分析中,改进算法在搜索精度、寻优稳定性和收敛速度上体现出了明显优势。

To address the problem that the search accuracy of grey wolf optimizer for function optimization is not high,an improved grey wolf optimizer is proposed using grouping optimization,Gaussian mutation and stochastic perturbation strategies.On the one hand,the information exchange between local search and global search is strengthened by using the grouping optimization strategy.On the other hand,the population diversity in evolution process is maintained by using Gauss mutation and stochastic perturbation strategies.Moreover,the population is updated with greedy idea.Simulation experiments demonstrate the effectiveness of the improved grey wolf algorithm by introducing multiple benchmark functions including single peak,multi-peak and fixed-dimension multi-peak.Compared with the other several advanced optimization algorithms,the improved algorithm has obvious advantages in search precision,stability and convergence speed.