Uncertainty Quantification in Integrated Fuel and Process Design

Abstract

Integrated fuel and process design aims to identify production-efficient renewable fuels that exhibit tailored combustion characteristics. In recent years, we have developed an optimization-based integrated fuel and process design framework that combines production pathway screening methods and predictive fuel property models. Inherently, the design of multicomponent fuels at an early stage of process development involves various sources of uncertainty that may affect the predicted economic and environmental performance, the optimal fuel composition, and the corresponding pathway selection. We present a systematic uncertainty analysis for the integrated design of multicomponent fuels and their production processes. To this end, we identify data and model uncertainties and quantify their influences on the integrated design of renewable drop-in fuels with Monte Carlo-based sampling methods. In addition, we evaluate the influence of structural uncertainty stemming from submodel choice within the optimization-based design approach. We show that the optimal fuel components and production pathways are quite robust to parameter variations and submodel choices. In contrast, the fuel component mass fractions and production costs are highly sensitive to uncertainties, underlining the importance of considering uncertainty effects in integrated fuel and process design. Such uncertainty-aware integrated fuel and process design can provide crucial guidance for future modeling and allocating research expenditures in fuel and process development.