Towards reduced order models of small-scale acoustically significant components in gas turbine combustion chambers

摘要: Gas turbine combustion chambers contain numerous smallscale features that help to dampen acoustic waves and alter the mode shapes. This damping helps alleviate problems such as thermoacoustic instabilities. During computational fluid dynamics simulations (CFD) of chambers, these small-scale are often neglected corresponding increase in mesh cell count augments significantly cost simulation while small physical size cells can present for stability solver. In where acoustics prevalent critical validity simulation, associated reduction overall cause with spurious, nonphysical noise prevents accurate transients limit cycle oscillations. Low-order dynamical systems (LODS) artificial neural networks (ANNs) proposed tested their ability represent a simple two-dimensional acoustically forced an orifice at multiple frequencies. These models were built using compressible CFD, OpenFOAM, placed between two ducts. The impedance has been computed multi-microphone method compared commonly used analytical model. Following this, flow field downstream modelled both LODS ANN Both methods have shown closely simulated flows much lower than original CFD simulation. Such may also assist flame quenching due cooling or design effusion cooled aerodynamic surfaces nozzle guide vanes (NGVs) blades.