This paper uses a divergence-conforming B-spline fluid discretization to
address the long-standing issue of poor mass conservation in immersed methods
for computational fluid–structure interaction (FSI) that represent the
influence of the structure as a forcing term in the fluid subproblem. We focus,
in particular, on the immersogeometric method developed in our earlier work,
analyze its convergence for linear model problems, then apply it to FSI analysis
of heart valves, using divergence-conforming B-splines to discretize the fluid
subproblem. Poor mass conservation can manifest as effective leakage of fluid
through thin solid barriers. This leakage disrupts the qualitative behavior of
FSI systems such as heart valves, which exist specifically to block flow.
Divergence-conforming discretizations can enforce mass conservation exactly,
avoiding this problem. To demonstrate the practical utility of immersogeometric
FSI analysis with divergence-conforming B-splines, we use the methods described
in this paper to construct and evaluate a computational model of an in
vitro experiment that pumps water through an artificial valve.