Cuckoo algorithm based on fuzzy C-mean value cluster

• Main Authors: GAO JING, WEI ZHIHONG, DONG ZEQUAN, YU LIJUN, ZHANG XUE, DING YING, WANG ZHENG'AN, HU YUKUN, WANG HUI
• Format: Patent
• Language: Chinese; English
• Published: 24.10.2017
• Subjects: CALCULATING; COMPUTING; COUNTING; HANDLING RECORD CARRIERS; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS

The invention belongs to the function optimization technical field, and specifically relates to a cuckoo algorithm based on a fuzzy C-mean value cluster; the fuzzy C-mean value algorithm is inputted in preference random wandering of the cuckoo algorithm, thus gathering components with the same property in the solution into one classification and updating same, and further enhancing the inter-dimension anti-interference capability; the step updating scope of the preference random wandering can be modified, thus increasing the searching directions, and improving the algorithmic diversity; the cuckoo algorithm can effectively improve the CS algorithmic convergence speed and improve the solution quality, and be applied to the high dimension function optimization solving problems. 本发明属于涉及函数优化技术领域，具体涉及种基于模糊C-均值聚类的布谷鸟算法。本发明通过在布谷鸟算法的偏好随机游动中引入模糊C-均值算法，将解中具有相同性质的分量集中在类进行更新，增强了维间抗干扰能力。同时更改偏好随机游动的步长更新范围，增加了搜索方向，提高了算法的多样性。本发明能够有效地提高CS算法的收敛速度并改善解的质量，尤其体现在求解高维函数优化问题上。