A Novel Deep Learning Approach for Tropical Cyclone Path Prediction Based on LSTM-BiLSTM

Abstract

Under global climate change, the frequency of typhoons and their strong wind, heavy rain, and storm surge increase, seriously threatening the life and property of human society. However, traditional tropical cyclone track prediction methods have difficulties in processing large amounts of complex data in terms of prediction efficiency and accuracy. Recently, deep learning methods have shown a potential capability to process complex data efficiently and accurately. In this paper, we propose a novel data-driven approach based on combination of Long Short Term Memory + Bidirectional Long short Term Memory (LSTM-BiLSTM) models to forecast tropical cyclone landing locations using the historical tropical cyclone tracks and various meteorological attributes. The features obtained from Finite Volume Method (FVM) is given to LSTM-BiLSTM. It is noted that FVM is integrated with a three-dimensional conservation of energy paradigm showcasing its effectiveness in capturing vital atmospheric and oceanic interactions. The model is trained on a real-world tropical cyclone dataset from the years 2007–2017. Through a comparison with existing forecasting methods, the results verified that our proposed model performed around 97% of accuracy, 93% of precision, 98% of recall, 2.66 e^–2 of error with 190.9 seconds of elapsed time for testing.