摘要
针对常规分形插值方法中垂直压缩因子参数无公认选择标准和有效途径的问题,提出一种分形插值的改进算法,用遗传算法对分形插值模型的垂直压缩参数进行了优化选择并进行了ARGO海温资料插值加密的对比试验。试验结果表明,经遗传优化后的分形插值方法在海温复杂细节结构和小尺度特征描述等方面较常规插值等方法更具优势,改进了分形插值结果的客观性和正确性。
Since there was no common recognized criterion and effective means for vertical compression factor in general fractal interpolation,the genetic algorithm was introduced to optimize the vertical compression factor in fractal interpolation model in this paper.Comparison experiments were carried out in ARGO sea surface temperature grid data fields.The results show that the optimized fractal interpolation method based on the genetic algorithm is better in describing complicated and detailed structure as well as micro-scale characteristics of sea surface temperature compared with other interpolation methods,which improves the objectivity and accuracy of fractal interpolation.
出处
《大气科学学报》
CSCD
北大核心
2010年第2期186-192,共7页
Transactions of Atmospheric Sciences
基金
国家重大基础研究计划项目(2007CB816003)
关键词
分形插值
垂直压缩因子
遗传算法
参数优化
ARGO资料
fractal interpolation
vertical compression factor
genetic algorithm
parameter optimization
ARGO data
