Lien, Hao-Pin (2023)
Spray-wall-flow interaction within a gasoline direct-injection (GDI) engine using Large Eddy Simulation.
Technische Universität Darmstadt
doi: 10.26083/tuprints-00026430
Ph.D. Thesis, Primary publication, Publisher's Version
Text
PhD_Thesis_final_Lien_HaoPin.pdf Copyright Information: CC BY-SA 4.0 International - Creative Commons, Attribution ShareAlike. Download (76MB) |
Item Type: | Ph.D. Thesis | ||||
---|---|---|---|---|---|
Type of entry: | Primary publication | ||||
Title: | Spray-wall-flow interaction within a gasoline direct-injection (GDI) engine using Large Eddy Simulation | ||||
Language: | English | ||||
Referees: | Hasse, Prof. Dr. Christian ; Lucchini, Prof. Dr. Tommaso | ||||
Date: | 19 December 2023 | ||||
Place of Publication: | Darmstadt | ||||
Collation: | vi, 160 Seiten | ||||
Date of oral examination: | 8 November 2023 | ||||
DOI: | 10.26083/tuprints-00026430 | ||||
Abstract: | With the need and urgency to the imposed regulation of zero-carbon emissions, the development of high-fidelity models for gasoline direct-injection (GDI) spray becomes crucial. This study first focuses on the development of robust Lagrangian models and a comprehensive exploration of the underlying physics across various operating conditions in a constant-volume chamber, ranging from early- to late-injection conditions. The Lagrangian models are extended to assess the spray-wall-flow interaction within an engine flow bench, simulating early-injection conditions of real GDI engines. The concept of these models is based on a Direct Numerical Simulation (DNS) inner-nozzle flow simulation, indicating that the liquid spray experiences complete atomization near the injector hole. Consequently, the deformation and secondary breakup of liquid droplets play a significant role in spray evolution. The models’ effectiveness and accuracy are meticulously validated against experimental data, including liquid and vapor phases obtained by diffuse back-illumination (DBI) and Schlieren measurements, respectively. An important aspect of the research involves the investigation of different droplet distribution models. Using the blob method, assuming the ejected droplet size equivalent to the injector diameter, is able to accurately capture global properties like liquid penetration length. However, it tends to cause delayed evaporation and breakup, resulting in an unphysical sharp plume tip downstream. To address future fuel-blended gasoline and E-fuels scenarios, the models have been extended to handle multi-component fuels. The successful simulation of a three-component gasoline surrogate (E00) demonstrates the models’ capability to reproduce both the overall spray plume characteristics and the spatial distribution of high- and low-volatile fuels. Furthermore, the research expands into the intricate spray-wall interaction within a constant-volume chamber under simulated cold start conditions. The simulation successfully replicates characteristic flows, such as wall jets and wall jet vortices induced by spray-wall interaction. Additionally, the phenomenon of spray cooling, resulting from air-entrainment-induced evaporation, is accurately reproduced. The simulated temperatures align closely with 0-D analytical results, exhibiting a temperature drop of about 20 K from its initial value. Although the simulation over-predicts heat transfer from the wall due to the constant temperature boundary condition, it matches the experimental aggregate wall film thickness data on the target wall, 40 mm from the injector tip. To comprehensively examine the spray-wall-flow interaction within a GDI engine, understanding the in-cylinder flow during the intake phase is imperative. Hence, a wall-resolved Large Eddy Simulation (LES) approach is employed to investigate free-stream and near-wall turbulence within an engine flow bench, simplifying the inherent complexity of the engine flow and focusing on the intake flow. The simulated in-cylinder large-scale motion and turbulence structure aligns well with reference experimental particle image velocimetry (PIV) data. Turbulence anisotropy analysis reveals a strong orientation toward axisymmetric expansion and contraction, respectively, attributed to the specific topological pattern of the engine flow characterized by the tumble vortex and the intake overflow jet. Moreover, the near-wall budget analysis facilitates investigating near-wall non-equilibrium effects, with a particular focus on the intake valve and liner wall region. The effects of the pressure gradient induced by the high Reynolds number intake flow are found to vary across different regions, suggesting that the classical wall function modeling approach based on the classical zero pressure gradient boundary layer may no longer be valid in internal combustion engine (ICEs) applications. Finally, the knowledge gained from the study is applied to assess the spray-wall-flow interaction in an engine flow bench under various mass flow rates (MFRs). As MFRs increase, the spray-flow interaction intensifies, and the heterogeneous behavior of all spray plumes becomes apparent. Plumes oriented along the intake flow jet exhibit higher penetration and lower evaporation, while those not aligned with the intake jet stream exhibit increased evaporation and reduced penetration. This observation confirms the significant impact of air entrainment induced by the intake flow on both evaporation and penetration length. Additionally, wall wetting is observed on the intake valves, and convective evaporation effectively reduces the fuel film, cutting its residual mass by up to 50% compared to the no-flow case when the mass flow rate is 100%. Under early-injection conditions, although the global turbulence kinetic energy experiences a transient increase during the injection, it eventually returns to its original values. |
||||
Alternative Abstract: |
|
||||
Status: | Publisher's Version | ||||
URN: | urn:nbn:de:tuda-tuprints-264302 | ||||
Classification DDC: | 600 Technology, medicine, applied sciences > 620 Engineering and machine engineering | ||||
Divisions: | 16 Department of Mechanical Engineering > Simulation of reactive Thermo-Fluid Systems (STFS) | ||||
Date Deposited: | 19 Dec 2023 13:36 | ||||
Last Modified: | 09 Feb 2024 11:52 | ||||
URI: | https://tuprints.ulb.tu-darmstadt.de/id/eprint/26430 | ||||
PPN: | 515410896 | ||||
Export: |
View Item |