Characterization and Modeling of Synthetic Jet Flow Fields
Abstract
The purpose of this research was to develop a neural network-based, lumped deterministic source term (LDST) approximation module for modeling synthetic jets in large-scale CFD calculations. The LDST approximation technique developed by the author and his students was employed. A series of two-dimensional calculations were performed to characterize unsteady single and multiple synthetic jet flow fields. The LDSTs were parameterized for different geometries as well as various flow conditions. These results were then used to train a neural network to approximate the LDSTs over a wide variety of conditions. Results include the successful development of a neural network LDST module for single two-dimensional synthetic jets for different Mach number flows and different jet orifice angles and characterization studies of both single and dual two-dimensional synthetic jets.
Document Details
- Document Type
- Technical Report
- Publication Date
- Apr 19, 2005
- Accession Number
- ADA433145
Entities
People
- Claudio Filz
- Daniel J. Lee
- Gregory Workman
- Katherine Grendell
- Matteo Pes
- Paul D. Orkwis
- Terry Daviaux
Organizations
- University of Cincinnati