Sensitivity studies of ANFIS based force recovery technique towards prediction of aerodynamic load

Abstract

Accelerometer force balance (AFB), in conjunction with various force prediction methodologies, has been used in hypersonic flow regime. Present studies focus on its ability to capture small and large alterations in the force and moment coefficients. For this objective, shock tunnel experiments are conducted on a bi-conic model without and with hemispherical or sharp aerospike. A three component AFB is equipped to measure acceleration responses during the experiments. Adaptive Neuro Fuzzy Inference System (ANFIS) is employed herein for the force recovery. Aerodynamic coefficients obtained from ANFIS are noticed to have encouraging match with the AFB theory. It has been shown here that, AFB and recently developed ANFIS based force recovery methodology can precisely predict as small as 6.3% and as large as 38% change in the force along with its temporal variation. Hence, this study strengthens the use of a soft computing technique like ANFIS for predicting steady state and temporal variations of aerodynamic coefficients during shock tunnel testing of hypersonic configurations.