CFD-driven neural network and SVM models for airfoil aerodynamic enhancement with bio-inspired riblets and plasma...

Enhancing the airfoil performance using advanced flow control techniques is an important challenge in the aerospace and automotive industries. In this article a hybrid computational and artificial intelligence (AI) approach is developed for aerodynamic performance enhancement of a NACA4412 airfoil by using a combination of bio inspired riblets and active plasma control. Semi circular, triangular and fillet riblets were tested with alternating current dielectric barrier discharge (AC-DBD) plasma devices and simulations were performed using computational fluid dynamics (CFD) technique at a Reynolds number of [Formula: see text] with angles of attack ranging from [Formula: see text] to [Formula: see text]. The performance assessment parameters considered in this work include lift coefficient [Formula: see text], drag coefficient [Formula: see text], lift to drag ratio ([Formula: see text]) ratio and vorticity contours in order to infer flow stability and efficiency. The simulation results indicated a reduced drag coefficient using riblets due to modifications in wall boundary layers along with a subsequent addition of AC-DBD devices to suppress vortex formation, delay separation and promote higher lift. Among all designs the bio-inspired fillet riblet with plasma actuator recorded maximum improvements with respect to [Formula: see text] with an increase of [Formula: see text] and [Formula: see text] in [Formula: see text] ratio with respect to base configuration. Additionally, to facilitate model prediction and simulation the machine learning algorithms were established using artificial neural networks with scaled conjugate gradient learning algorithm and support vector machine regressions. The results obtained indicated excellent model accuracy with high correlation coefficient and low mean squared error measurements using [Formula: see text] and [Formula: see text] predictions. These findings highlight the dual role of CFD and AI in the optimization of flow control strategies and confirm that the integration of physics based simulations with machine learning offers an efficient and reliable pathway for prediction and design innovation in aerodynamics.