Computational study of the Time Relaxation Model with high order deconvolution operator

This paper presents a computational investigation for a time relaxation regularization of Navier–Stokes equations known as Time Relaxation Model, TRM, and its corresponding sensitivity equations. The model generates a regularization based on both filtering and deconvolution. We discretize the equations of TRM and the corresponding sensitivity equations using finite element in space and Crank–Nicolson in time. The step problem and the shear layer roll-up benchmark is used to computationally test the performance of TRM across different orders of deconvolution operator as well as the sensitivity of the shear layer computations of the model with respect to the variation of time relaxation parameter in those cases.