PV4C

Motivation

The energy transition is a significant focus of European energy policy, driven by geopolitical factors and the need to reduce CO2 emissions. Austria, like many European countries, is actively transitioning to renewable energy sources such as photovoltaic (PV) systems. This transition is critical for reducing dependence on (Russian) fossil fuels and meeting EU REPowerEU goals.

Problem statement

However, there are several challenges in this transition:

• Temporal variability of PV systems: PV energy generation fluctuates due to weather conditions and shading, making it challenging to integrate renewable energy consistently.
• Coordination of generation and consumption: Energy generation patterns often do not align with consumption, requiring energy storage and transportation solutions.
• Network losses and expensive procurement: Transmission losses in the power grid lead to high energy costs and strain the network infrastructure.
• Grid expansion challenges: Labor shortages, material scarcity, and approval processes hinder the expansion of renewable energies.

Goals and innovation

The main objective of the project is to research technologies and methods to address these challenges and optimize the use of green energy, contributing to the Green Deal.

The research approach involves using satellite data, weather data, PV system locations, and historical generation data to develop models for identifying installed PV capacity and predicting PV power production. In contrast to smart meters, satellite-based forecasting allows the coordination of generation and consumption, thus relieving the grid infrastructure and allowing the full utilization of green energy.

These models will be integrated into a demonstrator tool to support energy communities, cities, and regions in optimizing PV electricity usage and increasing energy efficiency. The research objectives include AI-powered identification of PV modules, accurate solar and PV production forecasts, near real-time PV production estimation, and implementation of prediction results in a customer presentation tool. Expected results include improved large-scale identification of PV modules based on satellite data, accurate solar and PV production forecasts, downscaled solar forecasts, and enhanced nowcasting for PV production.

Overall, the integration of AI and renewable energies offers a promising solution to the challenges facing the energy sector, making the energy transition more efficient and sustainable while reducing dependence on fossil fuels.

This is done together with our partners energyfamily and GeoSphere Austria.