Multi-fidelity surrogate modeling and data fusion for accurate structural buffet load prediction

The overall goal of the project is to develop a mathematically rigorous multi-fidelity surrogate modeling (MFSM) and data analytics tool to fuse data from sources of varying facility (computations and flight tests) for accurate aircraft structural buffet load prediction. The salient aspects of the proposed solution are: (1) an adaptive framework to systematically tackle fusion and consolidation of buffet loading data of distinct availability from different sources, (2) improved accuracy in determining nonlinear buffet loading by bridging the resolution and fidelity among all available data, (3) identification of primary factors contributing to load generation, and (4) modular software architecture to automate the load prediction process and facilitate seamless integration into the Navy buffet loading workflow. In Phase I, all key software modules within the MFSM tool will be developed. Feasibility and capability will be demonstrated via case studies of Navy interest. Phase II efforts will focus on capability extension, model optimization, visualization and interaction, and extensive technology validation and insertion into Navy buffet load prediction workflow.