Search Results

A Machine Learning-Genetic Algorithm (ML-GA) approach was developed to virtually discover optimum designs using training data generated from multi-dimensional simulations. Machine learning (ML) presents a pathway to transform complex physical processes that occur in a combustion engine into compact informational processes. In the present work, a total of over 2000 sector-mesh computational fluid dynamics (CFD) simulations of a heavy-duty engine were performed. These were run concurrently on a supercomputer to reduce overall turnaround time. The engine being optimized was run on a low-octane (RON70) gasoline fuel under partially premixed compression ignition (PPCI) mode. A total of nine input parameters were varied, and the CFD simulation cases were generated by randomly sampling points from this nine-dimensional input space. These input parameters included fuel injection strategy, injector design, and various in-cylinder flow and thermodynamic conditions at intake valve closure (IVC).

The medium and heavy duty vehicle industry has fostered an increase in emissions research with the aim of reducing NOx while maintaining power output and thermal efficiency. This research describes a proof-of-concept numerical study conducted on a Caterpillar single-cylinder research engine. The target of the study is to reduce NOx by taking a unique approach to combustion air handling and utilizing enriched nitrogen and oxygen gas streams provided by Air Separation Membranes. A large set of test cases were initially carried out for closed-cycle situations to determine an appropriate set of operating conditions that are conducive for NOx reduction and gas diffusion properties. Several parameters - experimental and numerical, were considered. Experimental aspects, such as engine RPM, fuel injection pressure, start of injection, spray inclusion angle, and valve timings were considered for the parametric study.