Page 270 - AI for Good Innovate for Impact

AI for Good Innovate for Impact

Use Case 21: Driven Meteorological Modeling: Transforming

Wind Speed and Direction Prediction

Organization: Environmental Research Institute, University of Sadat City & SAMI Advanced
Electronics

Division: Meteorology and Environment

Country: Saudi Arabia

Contact Person(s):

Name: Mohamed Azouz
Email: azouzster@ gmail .com
Phone: +966504354224

1 Use Case Summary Table

Item Details
Category Energy Efficiency, Climate Action, Sustainability

Problem Addressed Traditional NWP models are computationally intensive and struggle to
deliver accurate, real-time wind forecasts – critical for renewable energy
and climate resilience.
Key Aspects of Solu- - AI-enhanced forecasting using LSTM, CNN, RNN
tion - Integration with ECMWF's AIFS
- Real-time and historical wind data from sensors, satellites, and
weather stations
- Improved grid stability and renewable energy output
Technology Keywords Artificial Intelligence, Machine Learning, Wind Prediction, Deep Learn-
ing, Smart Grid, AIFS, Meteorological Modeling

Data Availability Public and Private:
Public sources include ECMWF, NOAA, NASA, Copernicus
Private station and IoT sensor data

Metadata (Type of Time-series wind speed and direction data, satellite observations, real-
Data) time sensor feeds

Model Training and Deep learning models trained on multi-source wind datasets; integrated
Fine-Tuning with AIFS for global modeling robustness
Testbeds or Pilot Proof-of-concept under development; focus regions include the Red
Deployments Sea coast (Saudi Arabia) and Gulf of Suez (Egypt)

234