Time varying inflow generation using adjoint-based PDE constrained optimization for reactive flow simulations

This study provides a time-varying inflow generation methodology for high-fidelity numerical simulations. The present work utilizes the variational principles and adjoint-based partial differential equation (PDE) constrained optimization techniques. The optimization consists of two steps: forward integration and backward integration. The forward process integrates the given dynamical system, and the backward process integrates the adjoint state of the discrete dynamical system reverse in time. The current model-free approach provides an accurate time-varying inflow boundary condition for high-fidelity numerical simulations based on the experimental statistics. The proposed methodology is demonstrated for the highly three-dimensional swirl-dominated burner's jet and swirl inflow conditions.