基于LSTM和数值预报的机场短期能见度预报分析

Analysis of Airport Short-term Visibility Forecast Based on LSTM and Numerical Forecast

阐述利用WRF预报资料中的气象要素场,通过LSTM模型建立案例数据的能见度预报模型。为了评估气象要素场对模型的影响,通过相关性分析确定了能见度预报模型的最佳气象要素,探讨不同气象要素对能见度预测的影响。结果表明,随着输入气象要素的增加,LSTM模型在短期能见度预报方面可以取得较好效果。

This paper describes the meteorological element field in WRF forecast data, and establishes the visibility forecast model of case data through LSTM model. In order to evaluate the influence of meteorological element field on the model, the best meteorological element of visibility prediction model is determined through correlation analysis, and the influence of different meteorological elements on visibility prediction is discussed. The results show that with the increase of input meteorological elements, LSTM model can achieve good results in short-term visibility prediction.