Penalty formulation provides a stable and convergent approach for solving nonlinear magnetostatics problems.