摘要
采用实数编码的遗传算法,利用其良好的全局优化特点演化出各模式预报值在集成预报中的权重系数,得出集成预报的模型,实现了天气预报多种模式的集成处理,并利用天津气象科学研究所提供的实际测量数据进行了性能分析和验证.结果表明,利用遗传算法可以实现较好的集成性天气预报.
Using real-coded genetic algorithm and its good global optimization characteristics, weight coefficients of every model in the integrated forecasting were evolved. By using this method, integrated forecasting models were created and the integration of several models were realized. Finally the performance of the algorithm was analyzed and validated by real data provided by meteorological institute of Tianjin. The results indicate that the genetic algorithm can get a better integrated weather forecast.
出处
《天津科技大学学报》
CAS
2008年第4期80-84,共5页
Journal of Tianjin University of Science & Technology
基金
天津市科技发展计划项目(6YFSDSF04500)
关键词
集成天气预报
遗传算法
实数编码
权重
模式
integrated weather forecast
genetic algorithm
real-coded
weight
model
