Renewable fuels for compression ignition engines in various transport modes: effect of fuel properties on end-use performance

This doctoral dissertation belongs to the field of energy technology and focuses on the system level analysis of renewable liquid fuels for transportation purposes. Light-duty, heavy-duty and marine segments were in the scope of the study while compression ignition (CI) engine technology was the common powertrain. Based on statistical methods, the research linked engine performance with fuel properties, which has not been demonstrated yet. In this approach, data-driven black box modeling exploiting the multilinear regression method together with significance analysis and validations was successfully applied. The dissertation consists of four publications in peer-reviewed journals. Publication I concentrates on passenger cars and renewable drop-in fuels like hydrotreated vegetable oil (HVO) and traditional biodiesel (FAME). The experimental results from driving cycle test procedures are utilized in the model development. As an outcome, volumetric fuel consumption and carbon dioxide (CO2) emissions are predicted for light-duty fleet based exclusively on lower heating value, cetane number and density. In Publication II, the performance of heavy-duty vehicles is analyzed and predicted for various neat alternative fuels and their blends with fossil diesel. Simulations for new blending components are executed using heating value, density and cetane number. Publication III focuses on marine transport and suggests a model capable of predicting CO2 emissions from a medium-speed marine CI engine based on heating value, density and viscosity turned out to be significant properties. In Publication IV, the end-use challenges for the deployment of renewable fuels in the market are discussed and recommended interventions suggested. The novel approach presented in this doctoral dissertation enabled to link fuel properties with engine performance indicators like fuel consumption and tailpipe CO2 emissions. Three state-of-the-art models with high accuracy (coefficient of determination over 0.9 and all p-values below 0.05) were demonstrated in Publications I-III. The results from the modeling work represent the collective effect of fuel properties on end-use performance from the fleet perspective. Furthermore, all four publications support the deployment of renewable fuels in transportation by providing relevant modeling tools (Publications I-III) and recommendations (Publication IV) to decision makers. The outcomes of the dissertation serve for an unbiased comparison of the most promising fuel options that could be fitted to the existing infrastructure with minimum effort.