人工智能气象应用研究进展  

Progress in Research on Artificial Intelligence for Meteorology

出　　处：《中国气象科学研究院年报》2023年第1期71-75,共5页Annual Report of Cams

摘　　要：1人工智能在灾害天气预测分析中的应用1.4基于深度学习的融合降水临近预报方法及其在中国东部地区的应用研究为了实现对中国东部地区极端强降水的临近预报、预警,基于具有物理约束功能的PhyDNet构建了融合雷达反射率因子和分钟级降水观测资料的融合降水临近预报模型PhyDNet-RP,预测江苏省及其上游地区未来3h降水量,对比和探究了PhyDNet-RP、INCA(交叉相关外推+中尺度模式融合)、PhyDNet-P(仅包含降水资料)和UNet-RP(融合因子与PhyDNet-RP相同,但采用UNet模型)4种临近预报方法及对强降水增强过程的预测能力。1 Application of artificial itelligence in severe weather forecast and analysis 1.1 A study on loss function against data imbalance in deep learning correction of precipitation forecasts Sample imbalance has always been an important issue to be solved in heavy precipitation forecasts.Introducing appropriate constraints into the loss function of a machine learning model can mitigate the negative impact of sample imbalance on model training.This study explores the role of the Dice loss function in dealing with sample imbalance and verifies its application on the correction of heavy precipitation forecasts using the U-Net neural network.On this basis,the application of ordinal classification in forecast correction is further verified.The results show that the concept of Dice loss is highly similar to that of threat score.

关 键 词：临近预报 灾害天气 中尺度模式 人工智能 极端强降水 降水观测 雷达反射率因子 深度学习 

分 类 号：P45[天文地球—大气科学及气象学]