Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)

Legacy Department

Automotive Engineering

Committee Member

Dr. Zoran Filipi, Committee Chair

Committee Member

Dr. Robert Prucka

Committee Member

Dr. Mark Hoffman

Committee Member

Dr. Simona Onori


Ever tightening emissions and fuel economy regulations provide a strong impetus for research on high-efficiency low-emission engine concepts. In addition, CO2 emission regulation and energy security considerations motivate investigations focused on alternative fuels. In the current heavy-duty fleet, diesel engines dominate the market due to their unmatched thermal efficiency. Nevertheless, they suffer from NOx and soot emissions and require complex aftertreatment systems to meet stringent regulations. Due to recent advancements of the Natural Gas (NG) extraction technology, its supply has become increasingly abundant. Conversion of Heavy Duty engines to NG operation provides a most effective way of increasing utilization of this low-carbon fuel in transportation, and reducing its dependence on oil. Dual-fuel engines currently offered by OEMs are invariable conversions to spark ignited (SI) combustion. The compression ratio (CR) is lowered compared to diesel, and the engine operates with stoichiometric mixture in order to enable application of a Three-Way Catalyst. Dual fuel NG-diesel engines offer an attractive alternative. NG is mixed with air in the intake system, and a relatively small amount of diesel fuel is injected directly into the cylinder to initiate combustion. In that case, the conversion from a conventional diesel engine requires little modification of engine hardware. High CR is retained, and the engine can operate lean; hence, there is a prospect of achieving roughly the same thermal efficiency as in the case of a diesel baseline. Range anxiety is avoided, since the truck can continue running solely on diesel fuel if NG filling station is not available. Development of the dual-fuel concept requires systematic investigations of maximum substitution rates, while addressing challenges such as the combustion stability, knock, transient response and methane slip. Since combustion characteristics are not fully understood, and increased degree-of-freedom (DOF) in modern engines demand excessive calibration effort, traditional development process that relies on experimentation becomes very costly. A predictive engine-system simulation built around physics-based models can provide a paradigm shift, by enabling investigations of the design options and pre-development of the complex multi-variable control strategies on the computer. Extension of the physics-based approach can also yield very effective virtual sensing of intake charge flow, and support development of a next-generation transient air-to-fuel ratio control. Main contributions of this research are such models, namely (i) a hybrid diesel + NG dual-fuel combustion model, based on the multi-zonal diesel spray/combustion model and a turbulent flame propagation model of NG-air mixture, and (ii) a model of intake charge mass-flow rate that utilizes intake manifold pressure as a single pressure input, and simultaneously solves differential equations for gas flow and cylinder pressure. Dual-fuel combustion model addresses two modes of combustion taking place simultaneously in the cylinder. Diesel fuel injection, spray penetration, droplet evaporation, mixing, autoignition, heat release and emission formation are captured with a multi-zonal model. Original correlations were developed for ignition delay predictions in a dual fuel engine, spray penetration with high injection pressure, and heat release in the presence of Exhaust Gas Recirculation (EGR). Combustion of diesel fuel provides multiple ignition sites for the surrounding NG-air mixture. Initial flame kernels grow and merge to eventually form a flame front surrounding each of the sprays. An original model of the flame front geometry, and its interactions with surrounding flames and combustion chamber walls, is developed to provide foundation for application of the turbulent flame propagation model. Energy cascade approach is utilized for prediction of the time-based turbulent flow field characteristics, and a reduced chemical kinetics reaction mechanism is included for estimation of knock. Finally, sub-models are implemented in a Zero-D thermodynamic engine cycle simulation to create a predictive, and yet computationally efficient Quasi-D simulation tool. Accurate fuel metering requires accurate estimation of the intake charge mass. A universal feedforward intake air charge mass estimation method that requires reduced calibration effort, and a minimal set of sensors is pursued in this research. Simultaneous integration of differential equations capturing the variations of mass flow rate and the cylinder pressure yields a Single-pressure algorithm. It is capable of converging on correct values for both the mass air flow and cylinder pressure, given the known pressure upstream of the valve. Thus, it eliminates the need for the Mass Air Flow meter, and enables robust control of Air-to-Fuel mixture under both steady and transient operating conditions. Experiments in the engine test cell were utilized to aid model development and provide data for model validation. In case of the intake charge mass estimation, the newly developed Single-pressure model was implemented in the research-grade engine electronic control unit, to demonstrate its ability to provide accurate estimations over a federal driving schedule using an engine-in-the loop transient testing capability. Finally, rigorous in-vehicle testing was pursued subsequently to further test the accuracy, fidelity and real-time performance of the model.



To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.