Robust Data-Driven Aeroelastic Flight Envelope Tailoring
Abstract
The overall research objective was to develop and explore a dynamically data-driven decision support system to realize multi-objectively optimized system (joined-wing SensorCraft) stability under uncertainty. This project has resulted in a Dynamic Data Driven Application System (DDDAS) framework for a decision support system. This framework is composed of aeroelastic simulations, dynamic data-driven prediction, and operationally flexible robust optimization. Simulations are used to predict aeroelastic instabilities, which in turn are utilized in the prediction framework component to forecast the flutter boundaries for untested scenarios. An optimal design can then be constructed which best fits the predicted behaviors. Simulations are bolstered by the integration of the Fast Multipole Method (FMM) for computational reduction. Given that the aerodynamic calculations account for the majority of the computational cost, the aerodynamic calculations are accelerated through the FMM. A data-driven prediction framework was designed to include global, unsteady aeroelastic responses. In this framework, simulation data are integrated with measurement data from sparsely located sensors to make efficient and accurate predictions of unsteady responses of the system.
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 27, 2019
- Accession Number
- AD1085906
Entities
People
- Balakumar Balachandran
- Shapour Azarm
Organizations
- University of Maryland