A heterogeneous distributed prediction model for wind-solar energy production

This is the Marie Curie Individual Postdoctoral Fellowship project, which aims to bridge the gap in joint wind-solar energy forecasting by developing a Heterogeneous Distributed Prediction Model. With a focus on addressing challenges posed by data heterogeneity and siloed information, the project seeks to create a unified global model coordinated by a central server. Through four defined work packages, including model specification, client model development, model aggregation, and deployment/evaluation, the project aims to enhance forecasting accuracy. Emphasizing the importance of securing quality resources from the host university, the proposal aims to bolster scientific skills, foster innovation, and establish collaborative research networks. A two-way knowledge transfer approach is advocated to facilitate expertise exchange in energy big data and distributed modeling, aligning with the EU's climate objectives and contributing to advancements in renewable energy forecasting.

This project aims to improve predictions for combining wind and solar energy. It will develop a smart system that uses data from different wind farms and solar plants to make better forecasts, helping us use renewable energy more effectively and fight climate change.