The concurrent design of structures and material architectures at multiple intermediate length scales promises transformative capabilities for the future Air Force, ranging from integrated structural antennae to materials capable of sensing damage in real-time to performant joints for vehicles composed of complex components made of dissimilar materials. However, key challenges stand as barriers to this vision.

Though homogenization methods have long existed to bridge length scales with sufficient separation, methods to efficiently predict the behavior of material structures that defy length scale separations do not exist. Furthermore, damage tends to localize at a small scale and evolve into critical features at larger scales, which remains a seminal challenge for multiscale modeling to capture. Simulations are needed for the most extreme environments, ranging from hypersonic to space, requiring the incorporation of multiple physics at relevant scales. Additionally, multiscale predictions must quantify uncertainty across length scales, but new methods are required beyond the prohibitive Monte Carlo methods used today. Topology optimization algorithms offer a path to superior design, but algorithms must begin to be constrained by the manufacturing process. A key pillar of multiscale methods is varying approximations at each scale to capture pertinent phenomena, but robust methods to adaptively create suitable meshes and finite element approximations in the presence of complex topologies, geometric artifacts from preceding process simulations, and networks of cracks still need advancing. Finally, the high-performance computing hardware on which large-scale simulations rely is evolving rapidly, prompted by the opportunities created by the rise of machine learning, and demands methods to develop future-proof performance portable computational tools for emerging novel processing units.

Surmounting each of these challenges calls for ingenuity and the intersection of multiple disciplines. Candidates should have strong communication skills, embrace an interdisciplinary environment, and be an effective researcher with both attention to detail and a clear long-term vision.

key words

multiscale simulation; extended finite element methods; adaptive mesh refinement; manufacturing-constrained optimization; high-performance computing

Citizenship:  Open to U.S. citizens

Level:  Open to Postdoctoral and Senior applicants