Data-Driven Resolvent Analysis for Turbulent Flows with Explicit Nonlinearity Treatment using LANDO

Data-driven resolvent analysis (Herrmann et al., JFM, 2021) has been successfully applied to systems without significant nonlinearity, where it correctly identifies the most responsive forcing and receptive states. However, the extension of data-driven resolvent analysis to turbulent flows and multiphysics problems requires the separation of linear and nonlinear dynamics from data. We have previously explored this for small nonlinearity, i.e. transitional channel flow (Cao et al., APS DFD, 2024). In this work, we demonstrate the extension of data-driven resolvent analysis to turbulent flows by incorporating Linear and Nonlinear Disambiguation Optimization (LANDO, Baddoo et al., Proc. R. Soc., 2022) as a method to treat the nonlinearity. The proposed approach is demonstrated using minimal channel flow data for Re_τ 185.