Real-time Nonstationary Heteroscedastic Gaussian Process for Mean Flow Field Reconstruction
by
OFLG/402
Abstract:
Wind tunnel testing remains the gold-standard technique to validate the aerodynamic properties of any object, despite the significant progress in computational fluid dynamics. In wind tunnels, we typically sample the domain of interest using a robotic (or human) probe and reconstruct the continuous flow field using interpolation techniques. The operational costs of wind tunnels are enormous and reducing the sampling time is paramount to the efficient use of the facilities. We propose an active learning strategy paired with a sparse nonstationary heteroscedastic Gaussian Process regression algorithm to reduce the measurement time while providing an accurate mean flow field reconstruction. We propose a two-phase sampling strategy that combines an exploration phase, through a Metropolis-Hastings-inspired algorithm, and an exploitation phase, that relies on the estimated variance. The first phase ensures that we have enough samples to provide a good initial estimate and the second phase focuses on accurately estimating the heteroscedastic uncertainty. Numerical experiments on analytical fields show the efficiency of our approach, and real experiments performed in a large-scale wind tunnel facility validate it. The presented system allows for performing more efficient wind tunnel tests by reducing the measurement time and by providing accurate flow field estimations.
Laboratory for Simulation and Modelling LSM
SDSC Hub @ PSI