2024 | Ayyoub Frifra, Mohamed Maanan, Mehdi Maanan, Hassan Rhinane
This study explores the use of LSTM and XGBoost algorithms to predict storm characteristics and occurrence in Western France. The research combines data from a storm database and offshore buoy measurements from 1996 to 2020. The LSTM model was trained to predict temperature and pressure with high accuracy but struggled with extreme values for wave height and wind speed. The XGBoost model, on the other hand, performed exceptionally well in predicting storm occurrence. The study's innovative approach integrates both models to forecast storm characteristics and occurrence, expanding the scope to include a broader range of variables such as wave height, wave period, wind speed, temperature, pressure, and humidity. The results demonstrate the effectiveness of the combined model in reducing the impact of storms on humans and objects, highlighting its potential for enhancing preparation and mitigation strategies.This study explores the use of LSTM and XGBoost algorithms to predict storm characteristics and occurrence in Western France. The research combines data from a storm database and offshore buoy measurements from 1996 to 2020. The LSTM model was trained to predict temperature and pressure with high accuracy but struggled with extreme values for wave height and wind speed. The XGBoost model, on the other hand, performed exceptionally well in predicting storm occurrence. The study's innovative approach integrates both models to forecast storm characteristics and occurrence, expanding the scope to include a broader range of variables such as wave height, wave period, wind speed, temperature, pressure, and humidity. The results demonstrate the effectiveness of the combined model in reducing the impact of storms on humans and objects, highlighting its potential for enhancing preparation and mitigation strategies.