Numerical modeling of hydrogen mixing in a direct-injection engine fueled with gaseous hydrogen. (1st June 2023)

• Record Type: Journal Article
• Authors: Wu, Bifen; Torelli, Roberto; Pei, Yuanjiang
• Is Part Of: Fuel. Volume 341(2023)
• Publication Date: 2023-06-01
• Subjects: Hydrogen direct injection -- Hydrogen mixing -- Tumble ratio -- Turbulent Schmidt number; Fuel -- Periodicals; Coal -- Periodicals; Coal; Fuel; Periodicals; 662.6
• DOI: 10.1016/j.fuel.2023.127725
• Languages: English
• ISSNs: 0016-2361

Abstract

Highlights: Established a set of best practices for modeling a wide range of injection conditions and achieved accurate predictions of hydrogen mixing dynamics. Determined the correct geometry of the tumble plates installed in the intake ports to improve predictions of in-cylinder flowfields. Assessed relative importance of injection pressure and timing in determining fuel-air mixing and stratification. Abstract: Hydrogen is considered as one of the most promising options to achieve effective decarbonization of the energy and transportation sectors. As such, it has recently been receiving increasing attention because of its promising potential as an energy carrier for advanced energy and propulsion systems. With a focus on internal combustion engines, direct injection (DI) of gaseous hydrogen during the compression stroke offers great potential for high engine efficiency and specific power while reducing the risk of backfiring and pre-ignition issues. Therefore, many experimental and numerical efforts have recently been dedicated to understanding the physical and chemical behaviors of hydrogen in engine during mixing and combustion. This study focuses on computational fluid dynamics (CFD) modeling of the hydrogen DI process in a hydrogen optical research engine. Under the conditions studied, gaseous hydrogen is injected into the combustion chamber via a centrally located single-hole injector at a pressure of 100 bar. Two configurations, namely low- and high-tumble, are investigated to understand the impact of different in-cylinder flow patterns on the fuel-air mixture preparation. Simulations are carried out using the commercial CFD software CONVERGE. The in-cylinder turbulence is modeled with an unsteady Reynolds-averaged Navier-Stokes (URANS) formulation closed by the renormalization group (RNG) k - ε model. Several numerical methods and model constants, including but not limited to turbulent Schmidt number, are evaluated. The numerical results are systematically compared against experimental measurements of velocity and hydrogen concentration fields on the vertical center plane to assess the performance of the CFD model, unveil the physics of hydrogen mixing, and establish best practices for modeling hydrogen DI under relatively high injection pressure conditions.