Data-driven low-dimensional model of a sedimenting flexible fiber

The dynamics of flexible filaments entrained in flow are important for understanding many biological and industrial processes. This work describes a data-driven technique to create high-fidelity low-dimensional models of flexible fiber dynamics using machine learning; the technique is applied to sedimentation in a quiescent, viscous Newtonian fluid, using results from detailed simulations as the data set. Over a wide range of fiber flexibilities, the filament shape dynamics can be represented with high accuracy with only four degrees of freedom.