Accelerating CO2-Neutral Synthetic E-Fuels Development
Learn how AI-FIE is using deep learning-based AI algorithm to predict fuel sprays behavior and develop CO2-neutral synthetic e-fuels.
In the quest for more sustainable fuel solutions, the AI-FIE project led by City University of London, is focused on developing a deep learning AI algorithm designed to enhance the efficiency of predicting nozzle flow dynamics and spray patterns. This development aims to offer an alternative to the slower processes of traditional experiments and Computational Fluid Dynamics (CFD) simulations, although it builds upon these foundational methods. Central to the project is the integration of existing datasets with new data generated from tailored CFD simulations. This approach is intended to furnish the AI with a broad and nuanced understanding of various fuel properties and operational conditions, although it's acknowledged that the technology is still in its formative stages. During the initial phase of the project, which spanned the first two years, the team undertook the development of a thermodynamic model. This model, which aims to calculate the properties of different fuels including e-fuels and traditional fuel surrogates, is based on established theories such as the Perturbed-Chain Statistical Associating Fluid Theory (PC-SAFT) and Vapor-Liquid Equilibrium (VLE) calculations. While the model has been validated against a variety of data and has shown alignment with expected outcomes, it is recognized as one step in the ongoing journey of research and development in this area. Following the establishment of the thermodynamic model, the team developed a specific CFD code to aid in generating the necessary data for further AI training. This code is seen as a way to potentially reduce the need for extensive physical experimentation, thereby conserving resources and reducing waste. It's designed to simulate complex flows within fuel injection systems, a task it has performed with a degree of fidelity against established experimental data. However, the team remains cautious in their claims, recognizing the complexity of these systems and the challenges inherent in accurately modelling them. This UK-based research initiative, funded by the European Union's Horizon 2020 program under the Marie Sklodowska-Curie projects, represents a comprehensive approach to overcoming the complexities inherent in fuel technology development. By harmonizing Artificial Intelligence (AI), Computational Fluid Dynamics (CFD), and a thorough evaluation of existing experimental work, the project is poised to make a significant contribution to the wider field of energy research.
This project has received funding from the European Union's Horizon 2020 research and innovation program under the Marie Sklodowska-Curie projects No 101028449 (AI-FIE)