Reaction forces

[srosenbu]

I am not exactly sure what happens in the calculation of the reaction forces, but the current approach seems to be unnecessarily complex.

FenicsXConcrete/src/fenicsxconcrete/sensor_definition/reaction_force_sensor.py

Line 78 in 304d2b9

computed_force_x = -df.fem.assemble_scalar(df.fem.form(ufl.action(problem.residual, v_reac)))

Imo, it should be enough to sum up the nodal forces. This can be done on residual.vector.array directly

[joergfunger]

not sure if the residual vector actually contains the constrained dofs?