Feedback Circulation Control (CC) is a promising technology that improves the control of instabilities in flow fields. CC uses Reduced Order Models (ROMs) to simplify the high-dimensional equations in order to make it easier to work with them. In this project the modeling of the non-linear dynamics of the flow is done via ROMs based on Proper Or-thogonal Decomposition (POD) functions, these equations are obtained with the method of Galerkin Pojection (GP), which works with a ROM of the Navier-Stokes Equations (NSE), creating a suitable model for representing the separated and transient flow dynamics helped with the Computational Fluid Dynamics (CFD). This model is compared with 5 different methods: ANNs (Artificial Neural Networks), LSE (Linear Stochastic Estimation), mLSE (modified Linear Stochastic Estimation), QSE (Quadratic Stochastic Estimation) and mQSE (modified Quadratic Stochastic Estimation). These methods use data mapped from pressure and shear stress signals with separated and transient flow field to obtain the temporal coefficients which are compared with the original POD coefficients. The comparison of the pressure and shear stress correlation methods of the flow fields on an airfoil has been done with several sensors located equidistant on the flap surface. The obtained results demonstrate that ANNs and QSE are the best correlation methods. Then both methods have been optimized and finally QSE has been chosen as the strongest correlation compared to the other 4 methods, because of its reliability.