Abstract

This study reports the results of a combined experimental and numerical investigation of the performance of a three-dimensional film cooling slot configuration and its optimization using an evolutionary-based genetic algorithm. The primary objective of the optimization is to maximize the area-averaged film cooling effectiveness (ηad.avg) and minimize its standard deviation (ση) at a fixed coolant mass flux for a single liner. The experimental study is conducted on flow and heat transfer characteristics to validate the numerical model under laboratory conditions (low temperature and pressure). A design space is created using the Latin hypercube sampling technique, and it is solved using steady-state RANS simulations with the validated numerical model to estimate the ηad.avg and ση under actual engine conditions (high temperature and pressure). A surrogate model is developed using the kriging technique to predict the objective functions with the geometrical parameters of the slot. Following this, the optimum configuration is identified using the genetic algorithm. The geometrical parameters are slot jet diameter (d), slot jet pitch (p), lip taper angle (α), and lip length (L). The numerical results show that the optimum configuration outperforms the two reference configurations of the baseline combustor. At a blowing ratio (BR) of 1, the optimum configuration enhances the ηad.avg by 8% and reduces the ση by 6.6% compared to reference -1 configuration. Similarly, compared to reference -2, the optimum configuration enhances the ηad.avg by 19.8% and reduces the ση by 5.2.%. Furthermore, under actual engine conditions, a numerical study is conducted on a subsequent row of liners to determine the optimum length of the cooling ring. The numerical results show that the optimum slot configuration cools an additional liner length (G/S) of 5.5 compared to the reference-1 slot and results in a reduction of one pair of cooling rings in the combustor, which contributes to a 16.6% less coolant mass flux.

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