摘要
基于自然选择和自然基因机制的基因算法是一种优秀的寻优方法，它利用遗传操作算子来模拟生物界的优生劣汰的规律，是一种多路径全局优化方法。为增强经典基因算法对实际模型参数变化范围的适应性，加快其全局寻优速度和简化算法参数设置技术，本文给出了一种改进基因算法，并在渤海渔期预报、冰情预报中得到成功应用。该法可广泛用于各种工程模型优化之中。
The Genetic Algorithm(GA) is an excellent search procedure based on the mechanics of natural selection and genetics,which combines an artificial survival of the fittest with genetics operators abstracted from ecology.It can search for a global solution using the multiple paths.In this paper,an improved genetic algorithm named accelerative genetic algorithm (AGA) is presented to enhance the adaptability of GA to change range of parameters of practical models to accelerate global optimization and to simplify the configured technique of the parameters of GA.AGA is used very well to optimize the parameters of a model for forecasting fishing period of penaeus prawn in the Bohai Sea during overwintering migration.It was also successfully applied to forecast the interannual sea ice condition.It may be applied to optimize the different engineering models.
出处
《海洋环境科学》
CAS
CSCD
北大核心
1997年第4期7-12,共6页
Marine Environmental Science
基金
国家"九五"攻关课题
中国科学院资源与环境信息系统国家重点实验室资助
关键词
基因算法
海洋环境预报
genetic algorithm
superior species
method of optimization
marine environment forecasting
