Visibility Forecast for Airport Operations by LSTM Neural Network

Visibility forecast is a meteorological problems which has direct impact to daily lives. For instance, timely prediction of low visibility situations is very important for the safe operation in airports and highways. In this paper, we investigate the use of Long Short-Term Memory(LSTM) model to predict visibility. By adjusting the loss function and network structure, we optimize the original LSTM model to make it more suitable for practical applications, which is superior to previous models in short-term low visibility prediction. In addition, there is a ”time delay problem” when the number of hours time ahead we try to forecast becomes larger, this problem is persistent given the limited amount of available training data. We report our attempt of applying re-sampling to deal with the time delay problem, and we find that this method can improve the accuracy of visibility prediction, especially for the low visibility case.

[1]  Thomas G. Dietterich.  Machine Learning for Sequential Data: A Review , 2002, SSPR/SPR.

[2]  Peter Clark,et al.  Prediction of visibility and aerosol within the operational Met Office Unified Model. I: Model formulation and variational assimilation , 2008 .

[3]  Nan Wang,et al.  The Application of Deep Learning in Airport Visibility Forecast , 2017 .

[4]  J. Simon.  Resampling: The new statistics , 1995 .

[5]  Jürgen Schmidhuber,et al.  LSTM: A Search Space Odyssey , 2015, IEEE Transactions on Neural Networks and Learning Systems.

[6]  Afan Galih Salman,et al.  Weather Forecasting Using Merged Long Short-Term Memory Model (LSTM) and Autoregressive Integrated Moving Average (ARIMA) Model , 2018, J. Comput. Sci..

[7]  Miguel Rio,et al.  Internet Traffic Forecasting using Neural Networks , 2006, The 2006 IEEE International Joint Conference on Neural Network Proceedings.

[8]  Trevor Hastie,et al.  An Introduction to Statistical Learning , 2013, Springer Texts in Statistics.

[9]  Yifan Guo,et al.  Earthquake Prediction Based on Spatio-Temporal Data Mining: An LSTM Network Approach , 2020, IEEE Transactions on Emerging Topics in Computing.

[10]  Paulo Cortez,et al.  Internet Trafc Forecasting using Neural Networks , 2006 .

[11]  Gianluca Bontempi,et al.  Machine Learning Strategies for Time Series Forecasting , 2012, eBISS.

[12]  Qi Li,et al.  A Spatiotemporal Prediction Framework for Air Pollution Based on Deep RNN , 2017 .

[13]  Achim Zeileis,et al.  Forecasting Low-Visibility Procedure States with Tree-Based Statistical Methods , 2018, Pure and Applied Geophysics.

[14]  Jue Wang,et al.  Resampling Methods and Machine Learning: Resampling Methods and Machine Learning , 2009 .

[15]  Laxmidhar Behera,et al.  RNN Based Solar Radiation Forecasting Using Adaptive Learning Rate , 2013, SEMCCO.

[16]  Shotaro Minami,et al.  Predicting Equity Price with Corporate Action Events Using LSTM-RNN , 2018 .

[17]  Ayse Betül Oktay,et al.  Forecasting air pollutant indicator levels with geographic models 3 days in advance using neural networks , 2010, Expert Syst. Appl..

[18]  Xiang Li,et al.  Long short-term memory neural network for air pollutant concentration predictions: Method development and evaluation. , 2017, Environmental pollution.

[19]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.