Advanced Neural Network Modeling of Synthetic Jet Flow Fields
Abstract
The purpose of this research was to continue development of 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. The main exploration involved the grid sensitivity of the neural network model. A second task was originally planned on the portability of the approach to other solvers, but interesting developments on the first task prevented that study from being pursued.
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 01, 2006
- Accession Number
- ADA473581
Entities
People
- Paul D. Orkwis
- Terry Daviaux
Organizations
- University of Cincinnati