摘要
基于生态毒理动力学模型构造出可全局收敛的函数优化算法。在该算法中,将优化问题的搜索空间看成一个存在污染现象的环境系统,将一个试探解看成一个种群,采用生态毒理动力学模型对种群生长特征的变化规律进行描述。种群在污染作用下不断发生变化,能够抵抗住污染的强壮种群能够获得生长,而无法抵抗住污染的虚弱种群则停止生长。用环境和种群以及种群与种群之间的相互作用关系构造进化算子,这些算子从多种角度实现了种群之间的信息交换。因环境污染影响的是种群的很少部分特征,当种群演化时,只涉及到很少一部分种群特征参与运算,故提高了算法的收敛速度。测试结果表明本算法的精度和性能优于已有的群智能优化算法。
Based on ecotoxicology dynamics model we construct a function optimisation algorithm with global convergence. In the algorithm, the solution space of an optimisation problem （OP） is deemed as an environment system with pollution phenomenon, and a trial solution of OP is deemed as a population, the ecotoxicology dynamics model is used to describe the changes rule of some growth features of population. Populations constantly evolve under the effect of pollution; those strong populations who can endure pollution keep growing, while those week populations who cannot endure will stop growing. The interaction relations between environment and populations as well as among populations are used to construct the evolution operators; these operators realise the information exchange among populations in a variety of ways. Because the environment pollution gives influence on a very small part of features of populations, as they evolve, only a very small part of features take part in computation, it can substantially improve the convergence speed of algorithm. Test result shows that the algorithm outperforms existing population-based intelligent optimisation algorithms in both accuracy and performance.
出处
《计算机应用与软件》
CSCD
2015年第5期249-254,282,共7页
Computer Applications and Software
基金
陕西省科学技术研究发展计划项目(2013K11-17)
关键词
函数优化
智能优化计算
生态毒理动力学
环境污染
Function optimisation Intelligent optimisation computation Eeotoxieology dynamics Environment pollution
