Hydrogen combustion presents distinct challenges due to its high reactivity, which can lead to instability issues such as flashback, blow-off, and flame-turbulence interactions. To address these challenges, this approach uses FELiCS for linear stability analysis and adjoint-based data assimilation to predict and control flashback in hydrogen flames.
The primary focus is on optimizing combustor nozzle designs to prevent flashback and enhance flame stability. This is achieved by using FELiCS to perform linearized analysis of reactive, time-averaged flow fields, including velocity, temperature, and fuel composition. These fields are derived from high-fidelity simulations and assimilated experimental data. By applying linearized methods, including the adjoint approach, we can predict regions most susceptible to flashback and optimize nozzle shapes accordingly to ensure stable operation under varying combustion conditions. These designs are validated experimentally using additive manufacturing technologies.
This method offers a more efficient and systematic approach to combustor design compared to traditional trial-and-error methods, particularly in the highly reactive hydrogen combustion environment.