A Method for Solving Two-Stream Mixing using the Generalised Binomial-Langevin Multiple Mapping Conditioning Model
Paper
Paper/Presentation Title | A Method for Solving Two-Stream Mixing using the Generalised Binomial-Langevin Multiple Mapping Conditioning Model |
---|---|
Presentation Type | Paper |
Authors | du Preez, Matthew (Author), Wandel, Andrew P. (Author) and Lindstedt, R. Peter (Author) |
Editors | Klimekno, Alexander Y., Wandel, Andrew P., Lu, Yuanshen, Jacobs, Carolyn, Jahn, Ingo, Knibbe, Ruth, Rudolph, Victor, Shukla, Pradeep, Veeraragavan, Anand and Wheatley, Vincent |
Journal or Proceedings Title | Proceedings of the Australian Combustion Symposium 2021 |
ERA Conference ID | 42521 |
Number of Pages | 5 |
Year | 2021 |
Place of Publication | Toowoomba, Australia |
ISBN | 9780646854403 |
Web Address (URL) of Paper | https://www.anz-combustioninstitute.org/ACS2021 |
Conference/Event | Australian Combustion Symposium 2021 |
Australian Combustion Symposium | |
Event Details | Australian Combustion Symposium Rank C C C C C C C C C C C C C C |
Event Details | Australian Combustion Symposium 2021 Event Date 21 to end of 24 Nov 2021 Event Location Toowoomba, Australia |
Abstract | Newly-defined closures for the binomial-Langevin Multiple Mapping Conditioning (BLM-MMC) model are used to determine the proportion of stochastic particles that mix each time step. The parameter was previously treated as a constant. Two further developments were introduced to address the requirements associated with two-stream mixing configurations. First, the proportion of particles to mix is determined by requiring the MMC scalar variance to match the binomial-Langevin variance. To achieve this, particles are mixed using the Modified Curl’s model and a uniform random mixing amount until the scalar variance is lower than the binomial-Langevin variance. To exactly match the variance, the amount of mixing for the final particle pair is calculated via deterministic sampling from the distribution so compliance is guaranteed. Second, a standard Gaussian variable is introduced and defined so that the mapping function of the binomial Langevin scalar corresponds to its conditional mean. This conventional conditioning variable is introduced because its distribution is always continuous, whereas during the initial mixing period for two streams the probability density function of the scalar must be discontinuous, leading to segregated mixing. These changes are shown to correctly model non-reacting homogenous mixing of two distinct streams. |
Keywords | Multiple Mapping Conditioning, Two-Stream Mixing, binomial Langevin model, Curl’s Mixing |
ANZSRC Field of Research 2020 | 401204. Computational methods in fluid flow, heat and mass transfer (incl. computational fluid dynamics) |
401703. Energy generation, conversion and storage (excl. chemical and electrical) | |
400201. Automotive combustion and fuel engineering | |
Public Notes | Files associated with this item cannot be displayed due to copyright restrictions. |
Byline Affiliations | School of Mechanical and Electrical Engineering |
Imperial College of Science, Technology and Medicine, United Kingdom | |
Institution of Origin | University of Southern Queensland |
https://research.usq.edu.au/item/q7961/a-method-for-solving-two-stream-mixing-using-the-generalised-binomial-langevin-multiple-mapping-conditioning-model
67
total views4
total downloads2
views this month0
downloads this month