融合物理理解与模糊逻辑的分类强对流客观短期预报系统:(1)系统构成

Forecasting System for Short-Term Multi-Category Convective Phenomena Combining Physical Understanding and Fuzzy Logic PartⅠ:System Construction

提供准确的雷暴、短时强降水、雷暴大风和冰雹客观短期预报产品,对提高预报预警的预见期,及早采取有针对性的预防措施有重要意义。基于对四类强对流天气现象物理成因理解,给出了由国家气象中心牵头研发,融合模糊逻辑人工智能方法的分类强对流客观短期概率预报系统的流程框架和实现方法,详细介绍了该系统的结构特征,以及系统中用于雷暴、短时强降水、雷暴大风和冰雹四类强对流天气预报模型构建的关键预报因子、隶属度函数获取方法和权重因子配置等信息,并在此基础上探讨了物理理解与模糊逻辑人工智能相融合方法具有广泛适用性的本质,可以表征产生特定强对流天气现象的环境配置的多样性和复杂性。

Accurate and objective forecasts of thunderstorm,short-time severe rainfall,thunderstorm gale and hail are meaningful for extending the validation of warnings and taking targeted preventive measures.This paper introduces the framework and implementation ways of the objective forecasting system com-bining physical understanding and fuzzy logic artificial intelligence.This system,developed by the National Meteorological Centre(NMC),can provide short-term probability forecasts of thunderstorm,short-time severe rainfall,thunderstorm gale,and hail.The key predictors used for the four different convective weather phenomena,the methods for obtaining the membership functions,and the weighting sets of predictors are discussed.The property for the wide applicability of the combination method of physical understanding and fuzzy logic artificial intelligence is further investigated.It is concluded that the combination of the two can cover and reveal the key characteristics of the ever-changing environmental features favorable for a specific convective weather phenomenon.