A posteriori uncertainty quantification of PIV-based pressure data

Volumetric velocimetry for fluid flows

“…Similar approaches as in aerodynamic applications can be used, but with important differences: the flows of interest are "internal", and therefore the solid walls (and relative boundary conditions for integration) have a prominent impact on the entire domain; the viscous term in the momentum equation can rarely be neglected, and depending on the regime it can even be the dominant one (Schiavazzi et al 2017); and the inflow is usually oscillatory, making the relative importance of the various terms possibly time-dependent. In general, due to the multiple steps in deducing pressure from velocity, the analysis of uncertainty and its propagation is of paramount importance in both aerodynamic (Azijli et al 2016) and biomedical (Schiavazzi et al 2017) applications, and will likely remain a major line of research in the future.…”

Optimization of planar PIV-based pressure estimates in laminar and turbulent wakes

“…Recently, Blinde et al (2016) compared a number of pressure estimation techniques using synthetic data obtained from a zonal detached eddy simulation (ZDES) of an axisymmetric base flow, and showed the superiority of PTV-based material acceleration estimates for computing pressure fields, as well as the benefit of several techniques which implicitly correct the velocity field in the solution for pressure. Some recent studies have attempted to quantify the uncertainty in pressure estimations ( p ) given uncertainties in the velocity field ( u ) (Violato et al 2011;de Kat and van Oudheusden 2012;de Kat et al 2013;Laskari et al 2016;Azijli et al 2016), focusing on the Poisson equation problem.…”

Solenoidal filtering of volumetric velocity measurements using Gaussian process regression

“…Also, we are primarily interested in the posterior mean. Azijli et al (2015) use the posterior covariance to propagate the uncertainty from the measured velocity field through the Navier-Stokes equations, allowing a posteriori uncertainty quantification of PIV-derived pressure fields. The term R + HPH ′ will be referred to as the gain matrix A.…”