A Learning-Based Closed-Loop Fluid Flow Regulation

This paper presents a learning-based robust control strategy for fluid flow dynamic systems, designed to compensate for modeling uncertainties and unknown disturbances inherent to closed-loop active flow control applications. The method is grounded in nonlinear control theory and uses gradient descent learning rules to continuously update control parameters and disturbance estimate. The resulting closed-loop system achieves robust regulation while maintaining simplicity and interpretability in both implementation and analysis. Numerical simulations are conducted to validate the approach.