First database for operational concentrating solar power plant
- Concentrating solar power plants offer significant potential for sustainable energy generation, but face challenges, particularly with regard to the alignment of the heliostats.
- PAINT is the world’s first database of operational data from solar tower power plants, published jointly by researchers from the DLR Institute of Solar Research and the Karlsruhe Institute of Technology (KIT).
- This provides a unique basis for developing new AI methods and digital twins, as well as for testing power plant operations using simulation models.
Researchers from the DLR Institute of Solar Research and the Karlsruhe Institute of Technology (KIT) have published the first freely accessible dataset on the operation of the Solar Tower plant in Jülich, which is intended to accelerate research and development into solar thermal energy generation.
Solar tower power stations harness solar energy by generating heat rather than electricity directly. Heliostats – movable mirrors – precisely direct sunlight onto a receiver mounted on a tower. The heat generated there can be stored, used directly to generate electricity, or utilised for industrial processes. A distinctive feature is that these plants can also generate electricity on overcast days or at night, using stored heat, and can thus contribute to grid stability.
Although solar tower power stations are already commercially available, they are not widely used compared to photovoltaic systems. Until now, there has been a lack of data needed to test new methods for more efficient and reliable plants and to further develop the technology.
Accelerating the uptake of solar thermal technologies
“With PAINT, we are making real-world operational data from the Jülich Solar Tower openly available for the first time, thereby laying an important foundation for AI methods, digital twins and more efficient solar tower power plants. This is a significant step towards transferring solar thermal technologies more quickly from research into robust applications,” explains Robert Pitz-Paal, Director of the DLR Institute of Solar Research.
The PAINT database contains 849 gigabytes of operational and weather data from 2021 to 2024, more than 218,000 images of the heliostats, including details of the location, dimensions and orientation of the 2014 mirrors, as well as information on the condition of the mirrors.
The data has been published in accordance with the FAIR principles, ensuring that it is findable, accessible, interoperable and reusable. To this end, the research team is using the SpatioTemporal Asset Catalog (STAC) standard. This standard describes spatial and temporal data in a way that is readable by both humans and machines.
Testing power plant operations using a simulation model
In addition, the team provides Python software that allows researchers to download data for individual heliostats or specific time periods and integrate it directly into machine learning models. This makes it possible to extensively test and optimise power plant operations using the simulation model and to compare them objectively with approaches developed by other developers.
The database thus supports the development of digital twins, AI-based calibration methods, predictive maintenance and fault detection, as well as improved prediction of solar flux density. In future, further data from various plants could be added, thereby creating a common standard for open operational data in solar tower research.
PAINT emerged from work on ARTIST, an AI-supported, differentiable ray-tracing model for digital solar tower twins. The project involved researchers, engineers and technicians from the German Aerospace Center (DLR), the Karlsruhe Institute of Technology (KIT) and the Helmholtz AI platform.
Original publication
Kaleb Phipps, Mathias Kuhl, Marie Weiel, Marlene Busch, Jan Lewen, Nicolas Blumenröhr, Daniel Maldonado Quinto, Charlotte Debus, Felix Göhring, Oliver Kaufhold, Achim Streit, Robert Pitz-Paal, Markus Götz & Max Pargmann: The PAINT Database for Operational Concentrating Solar Power Plant Data Following FAIR Data Principles. Nature Energy, 2026. DOI 10.1038/s41560-026-02070-1.
More information: https://paint-database.org/