Optimization of slot geometry in shock wave boundary layer interaction phenomenon by using CFD–ANN–GA cycle

Abstract

Slot is one of the measures to control the Shock Wave–Boundary Layer Interaction (SBLI) used to avoid strong interference of shock waves with the boundary layer in supersonic flows. In this control measure, the Height of Triple Point (HTP) of λ shock significantly increases, compared to the one without controller, and cause a decline in shock power and pressure drops rate. In this paper, the main focus is on optimization of slot geometry as an influential parameter on the structure of the shock and flow characteristics by using Genetic Algorithm (GA). The averaged implicit Navier–Stokes equations and two equation standard k–ω turbulence models for the numerical simulation of the flow field have been used. The optimization problem is formulated in term of one objective function, namely, height of triple point maximization. Artificial neural network with two hidden layers has been used to achieve objective function based on the numerical simulation of the flow field data base. Root Mean Square Error (RMSE) was calculated for comparison and selecting the best algorithm in sequential steps. In order to simulate and compare the results with data obtained from experimental tests, the Cambridge University's wind tunnel tests and geometry have been used as the base design. The study demonstrates that, the HTP for optimized slot geometry is about 7.6 percentages more than base design.