Ir A.A. Verbeek
Room: HR N.213
MSc in Mechanical Engineering, University of Twente, Enschede, The Netherlands
Master thesis title: "Synthesis gas production by Pulsed Compression Technology"
Multi-scale modification of Swirling combustion for optimized gas Turbines
A popular way of generating heat and momentum (power) by combustion is to operate in a lean premixed regime. Lean combustion facilitates low NOx, low CO and low un-burnt hydrocarbons (UHC) emissions. Because of the excess of air, in this type of combustion, all fuel is properly burnt at a relatively low temperature. However, this way of combustion is hampered by issues of flame stabilization. The turbulent burning velocity should equal the local flame velocity at a proper and stable position. A turbulent burning velocity that is too high results in flash-back, while a burning velocity that is too low results in blow-off. In both cases destructive explosions might occur.
Low swirl concept
Conventionally stabilization of lean premixed combustion is obtained by application of high swirl. A post-flame recirculation pattern emerges (fig 2) in which hot products are continuously igniting the fresh gases.
|fig 1: EV Burner||fig 2: Recirculation zone|
A relative new and promising development is low swirl stabilization, which is based on flow divergence. This method has no recirculation and therefore only short residence time. This results in relative low emissions.
|fig 3: Low Swirl Burner||fig 4: Velocity field|
While swirl counters some of the stabilization problems of lean combustion, it also reduces the mixing near the flame. Therefore, perturbations that introduce small scale local turbulent wrinkling of the flame front is of key importance for the efficiency of the conversion of chemical energy. A proper balance should be struck between surface generation of small scales, and the associated local flame stretch. While extra flame surface is beneficial for combustion, local stretch tends to quench the flame. Thus a time mean flame brush with a certain thickness is established.
In recent years it was found that the mixing efficiency of turbulent flows could be strongly increased by the supply of the "correct" perturbations. This phenomenon is referred to as "resonant mixing"; it was shown that temporal and spatial perturbations at suitable scales increase mixing by 50% or more at the same power input.
The goal is to use the resonant mixing property in optimizing turbulent combustion systems in gas turbines. This requires the investigation and design of flow agitations under combustion conditions. This is all the more relevant since these systems are responsible for the majority of energy conversion on Earth. Several conflicting requirements come together: (i) low emissions, (ii) stable operations and (iii) complete combustion of fuel. To accommodate these requirements a multiscale approach is proposed in which lean premixed conditions should be combined with stabilizing large-scale swirl upon which "resonant" smaller scale flow perturbations are added to increase local flame wrinkling. To that end both practical and fundamental research is needed. Physical experiments are crucial to find out a priori the resonant regimes in actual combustors under realistic conditions, in which the low swirl stabilization concept works with good specifications. The experiments will provide data for validation of the findings of the numerical simulations.