Gradient Based Optimization and Control of Chaotic Multidisciplinary Systems via Least Squares Shadowing Adjoint Method
Abstract
This project has successfully developed the Least Squares Shadowing method for computing sensitivity gradients for optimizing the long-time behavior of chaotic dynamical systems, such as many unsteady flows. The Non-Intrusive-Least-Squares-Shadowing (NILSS) method has been developed theoretically. Its application to several computational fluid dynamics problems has been demonstrated.
Document Details
- Document Type
- Technical Report
- Publication Date
- Aug 10, 2018
- Accession Number
- AD1059547
Entities
People
- Qiqi Wang
Organizations
- Massachusetts Institute of Technology