Imagine a world where jet fuel comes from sunlight and waste, not oil. Sounds like science fiction, right? But what if I told you scientists are making it a reality, and they're using a 'digital twin' to do it faster and better? That's exactly what's happening at Synhelion's DAWN pilot plant.
Researchers at the Solar-Institute Jülich (SIJ) of FH Aachen, in collaboration with Synhelion engineers, have been working hard to create a virtual replica of Synhelion's groundbreaking solar fuels plant. This digital twin allows them to simulate the plant's operations in real-time, predict its behavior, and optimize its performance, all without having to experiment on the actual physical plant. The results of their work were presented in two insightful papers at SolarPACES 2025, showcasing the accuracy and reliability of these dynamic process models.
But what exactly is a solar fuels plant? At DAWN, concentrated solar power, harnessed by Synhelion's innovative technology, drives endothermic chemical reactions. These reactions transform a mixture of biogenic methane, carbon dioxide, and steam into synthesis gas, or "syngas." This syngas can then be further processed into liquid hydrocarbon fuels, including kerosene for airplanes and gasoline for cars.
And this is the part most people miss: Unlike traditional fuels derived from fossil sources, Synhelion's approach uses methane and carbon sourced from bio-waste. Furthermore, the heat needed for the process comes from concentrated sunlight, not by burning fossil fuels. This makes Synhelion a pioneer in creating a truly sustainable, "drop-in" replacement for the fuels we use today. Think of it: airplanes powered by sunshine and garbage!
The DAWN pilot plant, located in Jülich (and open for tours at the SolarPACES conference in the fall of 2026!), utilizes RED-certified sustainable biogenic waste as its methane and carbon source. It has already demonstrated the complete production chain under real solar operating conditions. Synhelion's solar-absorbing gas receiver is particularly noteworthy. Its unique design allows for extremely efficient conversion of solar irradiance from the heliostats, reaching temperatures as high as a mind-blowing 1500°C! But how does it achieve such scorching temperatures? Synhelion's proprietary design maximizes heat absorption and minimizes heat loss, allowing for unparalleled energy conversion efficiency.
Now, Synhelion is moving towards commercial production, and that's where the digital twin becomes indispensable. Plant operators need to adapt to changing conditions in real-time. Traditional physics-based models can be fast, but they often sacrifice accuracy. So, the team took a novel approach.
"We needed a system where we can offer simulation results in real time to aid the operators who run the plant and make the decisions," explained Falko Schneider, lead author from SIJ. "For example, when to switch operation modes, what kind of mass flows or compositions of the reactants to set. To achieve this, we need models that can deliver both high accuracy and high computational speed." He further elaborated that the model can also be used offline to test potential operating and control strategies.
The core idea is to replace the computationally intensive physics-based simulations with faster machine-learning models. The team started by creating a machine-learning model of the thermal energy storage, using synthetic data generated from the physics-based model. This model was then successfully validated against real-world data.
The overall system study validated three key components against real-time operation: the solar receiver, the reforming reactor (where the thermochemistry happens), and the thermal energy storage. The thermal energy storage is crucial because the thermochemistry must operate continuously, even when the sun isn't shining. Synhelion's high-temperature solid thermal energy storage system utilizes ceramic refractory bricks with internal channels to carry the heat transfer fluid.
The thermal energy storage model successfully reproduced charge-discharge cycles over 10 days, with errors of less than 3% in the upper storage layers. The surrogate model yielded results in milliseconds, compared to an average of over five seconds for the physics-based model – making it roughly 100 to 1000 times faster! On operational data from DAWN, it's even up to 50,000 times faster. This speed allows for many simulations to be run very quickly, making it practical for optimization tasks and real-time control.
Since the solar flux onto the receiver wasn't directly measured during operation, the researchers developed a simplified data-driven surrogate solar field model using a 4th-degree polynomial with ridge regression. The coupled surrogate-receiver simulation achieved outlet gas temperatures of approximately 1000°C, showing good agreement with experimental data.
But here's where it gets controversial... The model validation for the reforming reactor revealed major discrepancies, particularly under partial-load conditions. The model overestimated the reaction rate and heat consumption, with residual methane present, contrary to equilibrium predictions, and lower CO₂ conversion than expected.
"We still have some work to do there," Schneider admitted. The team realized that a thermochemical equilibrium wasn't being reached within the reactor, likely due to its relatively small size. They now need to implement kinetics, incorporating factors that account for the speed of reactants flowing through the catalyst.
And this is the part most people miss... Getting reliable measurements of the concentrated flux hitting the solar receiver is also a challenge. Without this input variable, validation of the receiver model becomes more difficult. The team is currently working on improving both the solar receiver and reforming reactor models.
Despite these challenges, the models provide a solid foundation for building a complete digital twin of the solar fuel process. The team is also exploring the use of machine-learning to accurately reproduce the complex chemistry involved.
This project is a collaborative effort between Synhelion, SIJ, and Germany's Institute of Future Fuels at DLR, with funding from the EU's TwinSF project. The long-term goal is to create a fully functional digital twin that enables real-time plant monitoring, predictive control, and virtual testing of control strategies before they are implemented on the actual solar fuels plant. Imagine being able to tweak the plant's operations in a virtual environment, optimizing its performance without risking any real-world disruptions!
Synhelion has already delivered its first solar aviation fuel to SWISS, marking a significant milestone in the quest for sustainable fuels. But the journey doesn't end there. The development of this digital twin is a crucial step towards scaling up solar fuel production and making it a viable alternative to fossil fuels.
What do you think about this innovative approach to creating sustainable fuels? Do you believe digital twins will revolutionize the energy industry? And what are the biggest challenges you see in scaling up solar fuel production? Share your thoughts in the comments below!
