ISBN 978-3-8439-4718-3

978-3-8439-4718-3, Reihe Ingenieurwissenschaften

Martin Pollack
Quadrature based moment methods for sprays and turbulent premixed combustion

332 Seiten, Dissertation Technische Universität Darmstadt (2020), Hardcover, A5

Zusammenfassung / Abstract

The optimization of several technical processes, such as combustion in engines or chemical multiphase reactions in process plants, requires a detailed knowledge and understanding of the phenomena involved and their interactions. Among the detailed measurement techniques available, computational fluid dynamics is considered an established and valuable methodology which provides detailed information about the system.

The simulation of many technical applications is, however, challenging, since large-scale processes, such as diesel and gasoline sprays in direct-injection systems, involve a huge amount of small-scale sub-processes, which cannot be directly resolved due to excessive numerical costs. Hence, these processes are commonly described statistically.

Moment methods are considered as an yet not fully established but very promising alternative to solve such statistical processes, since they provide several advantages.

The key point of this thesis is to develop and apply efficient and precise statistical moment-based descriptions of two processes appearing in combustion engines, namely spray formation and turbulent combustion. The closure for the moment transport equations in this work is provided by Quadrature-based Methods of Moment (QbMM), which are precise and numerically very efficient approaches.

In order to allow a more detailed evaluation of the spray physics, a novel, detailed, robust and numerically efficient algorithm has been developed to solve such processes for multivariate distributions.

The combustion approach in this work is generally based on the Transported

PDF method. Here, to provide a numerically more efficient solution strategy than the common PDF-resolving methods, the moment equations are derived and closed by applying QbMM combined with a tabulation approach.