Abstract

In the heavy-frame advanced turbine systems, steam is used as a coolant for turbine blade cooling. The concept of injecting mist into the impinging jets of steam was experimentally proved as an effective way of significantly enhancing the cooling effectiveness in the laboratory under low pressure and temperature conditions. However, whether mist/steam cooling is applicable under actual gas turbine operating conditions is still subject to further verification. Recognizing the difficulties of conducting experiments in an actual high-pressure, high-temperature working gas turbine, a simulation using a CFD model calibrated with laboratory data would be an opted approach. To this end, the present study conducts a CFD model calibration against the database of two experimental cases including a slot impinging jet and three rows of staggered impinging jets. Using the experimental results, the CFD model has been tuned by employing different turbulence models, computational cells, wall y+ values, and selection of near-wall functions. In addition, the effect of different forces (e.g. drag, thermophoretic, Brownian, and Saffman’s lift force) are also studied. None of the models are good predictors for all the flow regions from near the stagnation region to far-field downstream of the jets. Overall speaking, both the standard k-ε and RSM turbulence models perform better than other models. The RSM model has produced the closest results to the experimental data due to its capability of modeling the non-isotropic turbulence shear stresses in the 3-D impinging jet fields. For the 3-D flow fields, the nearest element from the wall must be set to approximately unity (y+ ≈ 1) to capture the correct flow structure. The simulated results showed that the calibrated CFD model could predict the heat transfer coefficient of steam-only case within 2 to 5% deviations from the experimental results for all the cases. When mist is employed, the prediction of wall temperatures is within 5% for a slot jet and within 10% for three-row jets.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call