【摘要】：針對花朵授粉算法易陷入局部極值、收斂速度慢的不足,提出一種具有入侵雜草策略的花朵授粉算法。該算法通過入侵雜草的繁殖、空間擴散和競爭策略,動態(tài)生成種群,增加種群的多樣性和有效性,使算法能有效地避免陷入局部最優(yōu),增強全局尋優(yōu)能力,提高收斂速度。通過8個CEC2005benchmark測試函數(shù)進行測試比較,仿真結果表明,改進算法的全局尋優(yōu)能力明顯優(yōu)于基本的花朵授粉算法、差分進化算法和蝙蝠算法,其收斂精度、收斂速度、魯棒性均較對比算法有較大提高。
[Abstract]:A flower pollination algorithm with invasive weed strategy is proposed to solve the problem that flower pollination algorithm is easy to fall into local extremum and converge slowly. The algorithm can dynamically generate population by invading weed propagation, spatial diffusion and competition strategy to increase the diversity and effectiveness of the population, so that the algorithm can effectively avoid falling into local optimum, enhance the ability of global optimization, and improve the convergence speed. Compared with eight CEC2005benchmark test functions, the simulation results show that the global optimization ability of the improved algorithm is obviously superior to that of the basic flower pollination algorithm, differential evolution algorithm and bat algorithm, and its convergence accuracy and convergence speed are better than those of the basic flower pollination algorithm, differential evolution algorithm and bat algorithm. The robustness is better than the contrast algorithm.
【作者單位】： 河池學院計算機與信息工程學院;江西財經大學信息管理學院;
【基金】：國家自然科學基金(61165015,61562032) 河池學院科研項目(XJ2015QN003)
【分類號】：TP18
，

本文編號：2270821