Code
01CD05224
Duration
01 September 2024 → 31 August 2025
Funding
Regional and community funding: Special Research Fund
Promotor
Research disciplines
-
Natural sciences
- Natural hazards
- Remote sensing
- Surface water hydrology
Keywords
neural network
dynamic interpolation
Flanders floods
Geo-AI
Flood prediction
Project description
In a study of flood events forecasting in Flanders, hybrid neural networks based models can complete and correct the measured data of measuring stations in the study area through dynamic interpolation, thereby predicting flood water depth more accurately. In addition, hybrid neural network based models could mimic the physical characteristics of physically based models, to improve the efficiency of flood forecasting.