Session: 06-06 Abnormal Combustion Modeling

Paper Number: 197007

197007 - Computational Modeling Framework for Characterizing Lubricant Oil Droplet Induced Preignition Events in Direct Injection Hydrogen Engines 

Abstract: 

Pre-ignition in direct-injection hydrogen (H₂) heavy-duty engines remains a critical challenge due to its sporadic occurrence and potential to induce severe abnormal combustion. This study presents a physics-based computational fluid dynamics (CFD) framework for capturing distinct oil-induced pre-ignition scenarios in a direct-injection H₂ heavy-duty engine. The modeling approach integrates oil-film dynamics, Euler-Lagrangian droplet transport, H₂–oil surrogate chemistry, and Reynolds-Averaged Navier–Stokes (RANS) turbulence modeling to investigate the mechanisms governing sporadic pre-ignition. Simulations show that pre-ignition predominantly originates in the exhaust-side squish region, driven by the anchoring of a low-velocity vortex core. This flow structure promotes preferential transport, accumulation, and evaporation of stripped oil droplets, resulting in localized oil-vapor-rich regions that are highly susceptible to autoignition. Distinct pre-ignition pathways are reproduced by coupling oil-film dynamics with moderate-temperature hot spots in the range of 800-1300K, demonstrating the importance of low-temperature oil chemistry in ignition initiation. The results further indicate that extremely high-temperature ignition sources are not required; moderately hot gas pockets interacting with oil-vapor-rich mixtures are sufficient to trigger early pre-ignition. Ignition is initiated primarily through oil vapor chemistry, while subsequent flame propagation preferentially follows hydrogen-rich regions due to the high reactivity and flame speed of hydrogen-air mixtures These findings identify the coupled roles of in-cylinder flow structures, oil-film dynamics, thermal hot spots, and mixture stratification in governing sporadic pre-ignition behavior in hydrogen engines. The proposed framework provides a predictive methodology for analyzing pre-ignition pathways and informing mitigation strategies for next-generation high-efficiency hydrogen combustion engines.

Presenting Author: Surya Kaundinya Oruganti Argonne National Laboratory

Presenting Author Biography: Surya Kaundinya Oruganti is a research scientist in the Department of Advanced Propulsion and Power at Argonne National Laboratory. His expertise is in high-fidelity computational modeling of ignition and turbulent combustion of both liquid and gaseous fuels.

Authors:

Surya Kaundinya Oruganti Argonne National Laboratory
Riccardo Scarcelli Argonne National Laboratory
Arun Ravindran Cummins Inc.
Sujith Sukumaran Cummins Inc.
Yu Zhang Cummins Inc.