基于深度学习的空气质量预报方法新进展

空气质量预报与人们的日常生活密切相关,其基本思想是分析历史空气质量数据,发现其内在的时空相关性,结合未来气象信息以及污染源排放量,对未来的空气质量进行预测。目前,环境管理和社会公众服务对空气质量预报提出了长时间、多维度、高精度的预测要求,一些新型的空气质量预测方法仍处于起步探索阶段。近年来,随着人工智能的普及与推广(特别是云计算与大数据的发展),深度学习这项基于传统人工神经网络的技术被国内外研究者所重视。笔者对现有典型的空气质量预报方法进行了阐述,包括数值预测模型方法、统计预报模型方法、基于机器学习模型的预测方法等,并重点介绍了该领域最新进展:基于深度学习模型的预测方法,并在此基础上进行了总结与展望。

New Progress for Air Quality Forecasting Methods Based on Deep Learning

Air quality forecasting is closely related to our daily lives.Its basic idea is to analyze the historical data of air quality,realizing the inner spatial-temporal correlation,combined with the future meteorological data as well as the pollution emissions,so that the future air quality can be predicted.Recently,environmental management and social public service have put forward some requirements for air quality forecast,such as long-time,multi-dimensional and high precision of prediction.Some new-type air quality forecasting methods are still in the preliminary exploration stage.In recent years,with the popularization and promotion of artificial intelligence,especially the development of cloud computing and big data,deep learning,a technology based on traditional artificial neural network,has been attached importance by researchers all over the world.In this paper,the existing typical air quality forecasting methods are described,including methods based on numerical prediction model,statistical prediction model and machine learning model.The latest progress in this field based on deep learning model prediction method is emphatically presented.On this basis,the summary of the existing typical air quality prediction methods and the development trend in the future are presented.