Krost, Philipp (2018)
Experimental Characterization and Quasi-Dimensional Modeling of Cyclic Combustion Variations in Spark Ignition Engines.
Technische Universität Darmstadt
Ph.D. Thesis, Primary publication
|
Text
180716_Krost_Dissertation.pdf - Accepted Version Copyright Information: CC BY-NC-ND 4.0 International - Creative Commons, Attribution NonCommercial, NoDerivs. Download (8MB) | Preview |
Item Type: | Ph.D. Thesis | ||||
---|---|---|---|---|---|
Type of entry: | Primary publication | ||||
Title: | Experimental Characterization and Quasi-Dimensional Modeling of Cyclic Combustion Variations in Spark Ignition Engines | ||||
Language: | English | ||||
Referees: | Hasse, Prof. Dr. Christian ; Beidl, Prof. Dr. Christian | ||||
Date: | 2018 | ||||
Place of Publication: | Darmstadt | ||||
Date of oral examination: | 13 June 2018 | ||||
Abstract: | Variances in spark ignition (SI) engine parameter settings are still expanding, thus the need for engine calibration is further increasing and virtual engine calibration to optimize engine parameters is being more frequently considered. In particular, the focus of engine calibration is to maximize fuel efficiency and power output while also reducing exhaust emissions up to the engine smoothness limit. This limit is determined by high cycle-to-cycle variations (CCV) and can be detected from indicated mean effective pressure (IMEP) fluctuations from one engine cycle to the other. These CCV dictate the stability of the combustion process and engine vibration; thus, the aim is to limit these variations to a certain comfort level. The objective of this thesis is to set up a zero-dimensional (0D) physical cyclic combustion variations model which can predictively describe CCV. In this work, first, extensive measurement data are produced by investigating five SI engines with different underlying combustion processes. These include both conventional engines and unconventional engines with a long expansion stroke via the crank and valve trains. Engine parameters are varied, in order to experimentally characterize CCV regarding the influence from the fluid mechanics, the chemical gas composition and the thermodynamical state. The CCV model itself is set up based on recently developed 0D models for turbulence, ignition and combustion; these are initially calibrated by means of 3D CFD data and measurement data. In the model development process, first, a stochastic model is developed to offer the possibility to impose fluctuations. From research into the literature, the physical causes of CCV are extracted. In particular, the new CCV model considers results from 3D CFD Large Eddy Simulations regarding the influence of global and local in-cylinder flow fluctuations, the most significant causes of CCV. For the first time within the 0D/1D simulation environment, flow fluctuations can be taken into account thanks to their availability in the 0D turbulence model used. In the first instance, it is shown that the new CCV model is able to reproduce experimentally observed CCV qualitatively. In order to also describe cyclic combustion variations quantitatively, beside the fluctuations due to these physical causes, factors influencing CCV, i.e. several engine parameters are introduced in the new CCV model. After the development process, the new CCV model is first verified with the design engine by means of experimental data and another commercial CCV model, considered state of the art. It is shown that the new CCV model is able to reproduce not only the fluctuations in the IMEP, but also the underlying fluctuations in the combustion process. Then, with no further calibration, the newly designed CCV model is successfully validated by means of the other engines and engine parameter variations investigated. Furthermore, it is explicitly shown that the new model, in contrast to the state-of-the-art model, is able to accurately describe CCV at two engine operating points with the same engine speed and load, but different internal residual gas rates and in-cylinder turbulence levels. Summing up, the new CCV model offers more encouraging results, enabling it to be used for virtual engine calibration even for cases in which no actual test engine is available. |
||||
Alternative Abstract: |
|
||||
URN: | urn:nbn:de:tuda-tuprints-76022 | ||||
Classification DDC: | 600 Technology, medicine, applied sciences > 620 Engineering and machine engineering | ||||
Divisions: | 16 Department of Mechanical Engineering | ||||
Date Deposited: | 27 Jul 2018 08:35 | ||||
Last Modified: | 09 Jul 2020 02:10 | ||||
URI: | https://tuprints.ulb.tu-darmstadt.de/id/eprint/7602 | ||||
PPN: | 434423289 | ||||
Export: |
View Item |