Artificial Intelligence Applied to Energy Systems
[Open soon]
Work description
The increasing penetration of renewable energy sources and the decentralization of energy systems make energy forecasting a critical component for the efficient operation of power systems and local electricity markets. The variability and uncertainty associated with renewable generation (e.g., solar and wind), as well as electricity demand and prices, require robust predictive models to support both operational and strategic decision-making. In this context, the proposed work plan focuses on the development and validation of advanced forecasting models, leveraging artificial intelligence approaches, including generative models and explainable AI techniques, aligned with the project objectives. The grantee will contribute to integrating these forecasts into decision-support tools for local electricity markets and distributed energy systems. The main planned activities are: - Reviewing the state of the art in energy forecasting, including statistical, machine learning, and deep learning approaches. - Developing forecasting models for renewable generation, electricity demand, and market prices. - Implementing advanced models (e.g., LSTM, Transformers, generative models) for time-series prediction. - Evaluating model performance and analysing forecasting uncertainty. - Integrating forecasting models into optimization and energy management frameworks for local electricity markets (LEM). - Applying Explainable AI (XAI) techniques to interpret predictive models. - Preparing technical reports, scientific papers, and dissemination materials in collaboration with the research team.
Academic Qualifications
Master in Electrotechnical Engineering or similar
Minimum profile required
- Basic knowledge of time-series analysis and data analysis;- Programming skills in Python;- Fundamental understanding of machine learning;- Fluency in English (written and spoken);
Preference factors
- Experience in energy time-series forecasting (renewable generation, demand, prices); - Knowledge of machine learning and deep learning applied to energy problems; - Experience in Python programming and use of libraries such as Pandas, Scikit-learn, TensorFlow, or PyTorch; - Familiarity with advanced forecasting models (e.g., LSTM, Transformers, probabilistic models); - Knowledge of electricity markets and energy systems; - Experience in data analysis and visualization; - Ability to develop models integrated with optimization and energy management;
Application Period
Since 01 May 2026 to 15 May 2026
[Open soon]
Centre
Power and Energy Systems