Resolving the incompatibility of hydrodynamics calculations between popular BEM solvers

Abstract

This paper compares different Boundary Element
Method (BEM) solvers used to calculate hydrodynamic
effects on floating bodies to reveal divergence in
their calculations due to different modeling conventions.
These differences become evident when analyzing different
wave headings, multiple bodies, or non-axisymmetric
floating bodies, i.e., situations where the pose, orientation,
and position of the body with respect to the global reference
frame and coordinate frame will influence the hydrodynamics
calculations. BEM-based solvers use the Linear
Potential Flow Theory (LPFT) which is a common midfidelity
modeling approach for offshore floating structures.
In this work we show that there is a lack of consensus in
the commonly used BEM solvers, viz., WAMIT, Capytaine,
NEMOH, AQWA, HAMS, and OrcaWave. We demonstrate
the differences in hydrodynamic coefficients calculated
by different BEM solvers by considering two triangular
prisms. The available documentation for the BEM solvers
at the time of writing, along with communications with
the developers of these BEM solvers left some ambiguities.
However, this work was able to reverse-engineer the cause
for the differences in the results in a systematic manner.
Furthermore, this work presents mathematical expressions
that can convert hydrodynamic coefficients from each of
the considered BEM solvers so as to enable verify of the
hydrodynamic calculations across different BEM solvers.
Finally, the findings were implemented in BEMRosetta
(https://github.com/BEMRosetta/BEMRosetta) so that the
general community can easily convert or verify their hydrodynamic
calculations.