Iterative symbolic regression for learning transport equations

Computational fluid dynamics (CFD) analysis is widely used in chemical engineering. Although CFD calculations are accurate, the computational cost associated with complex systems makes it difficult to obtain empirical equations between system variables. Here, we combine active learning (AL) and symbolic regression (SR) get a equation for variables from simulations. Gaussian process regression-based AL allows automated selection of by selecting most instructive points available range possible parameters. The results these experiments then passed SR find models. This approach scalable applicable any desired number design To demonstrate effectiveness, use this method two model systems. We recover an pressure drop bent pipe new predicting backflow heart valve under aortic insufficiency.