No abstract

No abstract

No abstract

The Sandia/ETH-Zurich CO/H 2 /N 2 non-premixed unconfined turbulent jet flame (named 'Flame A') is numerically simulated by solving the unsteady compressible reactive NavierStokes equations in a three-dimensional axisymmetric formulation, hence, in a formally twodimensional domain. The turbulent combustion closure model adopted is the Fractal Model, FM, developed as a subgrid scale model for Large Eddy Simulation. The fuel is injected from a straight circular tube and the corresponding Reynolds number is 16 700, while the air coflows. Since the thickness of the nozzle is 0.88 mm, and the injection velocity high, ∼104 m s −1 , capturing the stabilization mechanism of the actual flame requires high spatial resolution close to the injector. Results are first obtained on a coarse grid assuming a fast-chemistry approach for hydrogen oxidation and a single step mechanism for carbon monoxide oxidation. With this approach the flame is inevitably anchored. Then, to understand the actual flame stabilization a more complex chemical mechanism, including main radical species, is adopted. Since using this chemistry and the coarse grid of previous simulation the flame blows off numerically, attention is focused on understanding the actual flame stabilization mechanism by simulating a small spatial region close to the injection with a very fine grid. Then, analysing these results, an artificial anchoring mechanism is developed to be used in simulations of the whole flame with a coarse grid. Unsteady characteristics are shown and some averaged radial profiles for temperature and species are compared with experimental data.

No abstract

A progress variable/flame surface density/probability density function method has been employed for a Large Eddy Simulation of a CH 4 /Air turbulent premixed bluff body flame. In particular, both mean and variance of the progress variable are transported and subgrid spatially filtered gradient contributes to model the flame surface density (that introduces the effect of the subgrid flame reaction zone) and to presume a probability density function (that introduces the effect of subgrid fluctuations on chemistry). Chemistry is preliminarly tabulated in terms of laminar premixed flames and enthalpy is included as a new coordinate in their tabulation to take into account heat losses in the flowfield. Then, the PDF is used to build a turbulent flamelet library. The filtered mass, momentum, enthalpy and scalar equations mentioned above are integrated by an explicit scheme using finite differences, 2nd-order accurate in space and third order in time, over a cylindrical non-uniform grid using a staggered mesh. The bluff-body geometry is modelled by using the Immersed Boundary Method. The numerical predictions are compared with the available experimental data.

Although CSP has reached technological maturity, high capital investment and specific electricity cost remain the major development barriers. To reduce them, highly efficient, integrated, and cheaper CSP components are urgently needed. In this paper, we investigate a novel CSP plant configuration with a single-tank Thermal Energy Storage (TES) fully integrated with the steam generator. The objective of this research is twofold: i) provide a reliable model of single-tank thermal storages with integrated steam generator; ii) identify two optimized CSP plant designs to achieve best energetic and economic performances. To achieve these aims we developed a numerical model of the main system components and validated it against experimental data. This model was then integrated in a full simulation and heuristic design optimization of the plant. The results revealed that the system proposed can generate electricity in middle-Italy (Rome) at a cost of 230.25 $/MWh with a 15 % reduction compared to the double tank option. Furthermore, if cogeneration is used to recover the waste heat, this system is an interesting option for users such as small districts, university campuses and hospitals. In the latter case, the optimized system pays off in 6 years and covers 80 % of the heating and cooling requirements.

No abstract