摘要
建立了自主式粒子群优化模型,进一步完善了经典粒子群优化算法的学习机制,提高了粒子学习的自主性。在该模型的基础上,针对自主选择共享信息问题,提出了一种学习榜样自主获取的粒子群优化算法,该算法粒子依据自身的内在特征合理地选择学习榜样,充分地利用了进化过程中产生的信息,有效抑制共享信息的流速。对常用单峰多峰基准函数进行了测试,验证了该算法的效率和优越性。
The self-regulated model of particle swarm optimization was built to further consummate the learning mechanism of the classical particle swarm optimization and enhance the self-learning ability of the particle. The particle swarm optimization with self-regulated acquisition of the example was developed to actively select sharing information in the self-regulated model of particle swarm optimization. The particle can select example by the inherent factors, which makes full use of the beneficial information generated in evolution and effectively restrains velocity of the flow of the sharing information. The testing of uni/multi-modal benchmark functions commonly used in the evolutionary computation proves the efficiency and superiority of this algorithm.
出处
《重庆邮电大学学报（自然科学版）》
北大核心
2009年第4期507-511,共5页
Journal of Chongqing University of Posts and Telecommunications(Natural Science Edition)
基金
国家自然科学基金(60773113)
重庆市自然科学基金(2008BA2017)
关键词
自主式学习
粒子群
认知水平
个体差异
self-regulated learning
particle swarm
cognitive level
individual difference
