Get ready for a groundbreaking innovation that could revolutionize the way we think about energy and sustainability! Researchers have created a digital twin, a virtual replica, to optimize Synhelion's solar fuels plant, DAWN.
Collaborating with Synhelion's engineers, a team of solar researchers from the Solar-Institute Jülich (SIJ) presented their findings at SolarPACES 2025. They assessed the accuracy and reliability of dynamic process models for solar fuel production, utilizing data from Synhelion's DAWN pilot plant.
Synhelion's solar fuel plant, DAWN, is a game-changer. It uses concentrated solar power to drive chemical reactions, converting biogenic methane, carbon dioxide, and steam into synthesis gas (syngas). This syngas can then be transformed into liquid hydrocarbon fuels, such as kerosene for aviation or gasoline. The beauty lies in its sustainability; the methane and carbon sources are derived from bio-waste, and heat is generated from solar energy, eliminating the need for fossil fuels.
DAWN, located in Jülich, has demonstrated the entire production chain under real solar conditions. It utilizes RED-certified sustainable biogenic waste as its methane and carbon source, making it a pioneer in sustainable liquid fuel alternatives.
But here's where it gets controversial... Synhelion's solar-absorbing gas receiver is a unique design, capable of achieving temperatures as high as 1500°C. How do they do it? That's the million-dollar question!
As Synhelion begins commercial solar fuel production, real-time adaptation becomes crucial. Traditional physics-based models are fast but often lack accuracy. So, the research team took a bold approach.
Lead author Falko Schneider explains, "We needed a system to provide real-time simulation results to aid plant operators in making critical decisions. Our goal was to develop models that offer both high accuracy and computational speed."
The studies focused on validating physics-based and machine-learning models for simulating DAWN's operations. The team created a machine-learning model for the thermal energy storage, using synthetic data from the physics-based model, and successfully validated it.
In the overall system study, "Validation of a Dynamic Process Model of a Thermochemical System for Solar Fuel Production," all three key components were validated against real-time operation: the solar receiver, the reforming reactor, and the thermal energy storage. The latter received the most attention.
Synhelion's thermal energy storage system uses ceramic refractory bricks with internal channels for heat transfer. The team's surrogate model for the thermal storage reproduced charge-discharge transients over 8 cycles across 10 days with remarkable accuracy, achieving errors of less than 3% in temperature swing in the upper layers. However, deviations increased towards the bottom, likely due to material property variations and measurement uncertainties.
The surrogate model's performance was impressive, offering results in milliseconds compared to the physics-based model's average of over five seconds - a speed difference of up to 50,000 times! Its consistent run times make it ideal for optimization and real-time control tasks.
Since solar flux onto the receiver wasn't directly measured, the researchers developed a simplified data-driven surrogate solar field model using ridge regression. The coupled surrogate-receiver simulation achieved outlet gas temperatures of approximately 1000°C, closely matching experimental data. However, a minor systematic bias in the surrogate model led to slight temperature overestimation.
But the reforming reactor model revealed major discrepancies, especially under partial-load conditions. The model overestimated reaction rates and heat consumption, with residual methane contrary to equilibrium predictions and lower CO₂ conversion than expected. "We still have some work to do there," Schneider acknowledged.
The team plans to implement kinetics to address these issues. "We need to consider factors like the flow rate of reactants through the catalyst and how close it gets to a thermochemical equilibrium. These factors can be fitted with operational data from DAWN to improve the model's accuracy," Schneider explained.
The solar receiver and reforming reactor models require further refinement. The team aims to develop a detailed heliostat field model and implement a kinetic model or effectiveness factors to capture the system's behavior under partial loads more accurately.
"Our long-term goal is to develop a functioning digital twin that enables real-time plant monitoring, predictive control, and virtual testing of control strategies before implementation," Scheider emphasized.
This project, funded by Synhelion, SIJ, and Germany's Institute of Future Fuels at DLR, is part of the TwinSF project in the EU. It combines physical models with data-driven approaches to guide operations in alternative fuel production plants powered by concentrated solar energy.
And this is the part most people miss... The potential impact of this digital twin technology is immense. It could revolutionize the way we produce and control solar fuels, making sustainable energy more accessible and efficient. But what do you think? Is this technology the future of sustainable energy, or are there potential pitfalls we should consider? We'd love to hear your thoughts in the comments!
