Calculation of Uncertainty Propagation Using Adjoint Equations

A.K. Alekseev, I. M. Navon

The uncertainty of flow parameters depending on the error of input data (initial, boundary conditions, coefficients) may be efficiently calculated using adjoint equations. This approach is extremely effective for uncertainty estimation at certain checkpoint because it needs to solve only single (adjoint) system of equations above the system describing the flow-field. The fields of "adjoint "temperature", adjoint "density" etc. are used to calculate the transfer of uncertainty from all input data.