SS03: Advances in Machine Learning for Photovoltaic System Optimization and Control in Modern Energy Grids
Peter Glösekötter
FH Münster, Germany
Joseph Moerschell
Haute École Spécialisée de Suisse Occidentale, Switzerland
Ignacio Rojas
Universidad de Granada, Spain
Tilman Sanders
FH Münster, Germany
Markus Gregor
FH Münster, Germany
Sarah Trinschek
FH Münster, Germany
Catalin Stoean
University of Craiova, Romania
Abstract
With the increasing integration of renewable energy sources like photovoltaics (PV) into modern energy systems, achieving grid efficiency, while maintaining stability, have become complex challenges, especially due to the variable nature of PV power generation. Machine learning (ML) provides promising solutions for forecasting, monitoring, and optimizing energy flow and control to maximize PV utilization, but still ensuring grid stability. This session aims to bring together researchers and practitioners to discuss emerging ML techniques specifically designed for optimizing photovoltaic systems and sensors in the context of modern energy grids.
Topics of interest include (but are not limited to):
- PV production forecasting using time series analysis
- Sky-camera image processing
- Estimation of grid-connected storage state through ML
- Sensor placement optimization through machine learning
- Fault detection and stability analysis in power grids using ML
- Optimizing power systems through metaheuristics
- Explainable AI models for energy system management
- End-to-end ML pipelines for PV fault detection and maintenance planning