Possibility to have additional unknown vector solutions in Newton Raphson and Assembly subroutines
In the near future, new unknowns solutions need to be accounted for in addition to "displ" and "p_field", such as "Concentration" when dealing with Diffusion coupled problem.
I don't know if this needs to introduce a structure variable called "Solution", such that we no longer modify the number of inputs in phase_field.fem.solver.equilibrium.newton_raphson
and phase_field.fem.solver.pf.newton_raphson
; depending on the problem studied (for example elasticity and fracture only), one could call his solution vectors by Solution.displ and Solution.pf.
The Newton Raphson subroutine will look like this
function [STIFFNESS_MATRIX, strain_en_undgr, Solution, res_0, res] = newton_raphson(... sys, D_RHS_tract, Solution, label_el,... args... )
instead of
function [STIFFNESS_MATRIX, strain_en_undgr, displ, res_0, res] = newton_raphson(... sys, D_RHS_tract, p_field, displ, label_el,... args... )
Obviously, this new change means also modifying how the unknowns are called in every fortran subroutine...