Cigdem Kilic: Machine learning driven prediction of cold properties for lignocellulosic fuels

The global transition to low-carbon energy demands various solutions, with renewable lignocellulosic fuels being one key option. However, their widespread use as fuels demands both production technology maturity and proper physico-chemical properties, particularly operability in cold conditions. To address this issue, machine learning is applied to unravel the complex relationships between molecular structure and fuel properties. This research aims to develop predictive models to guide the formulation of chemically optimized, sustainable blends for reliable operation in real life engines.