Fast and accurate component models are essential for the development of thr
ee dimensional millimeter wave integrated circuits. In this paper, an accur
ate and efficient artificial neural network (ANN) model for the gap discont
inuity in shielded coplanar waveguide is presented. The ANN model is develo
ped with a set of training data given by the finite-difference time-domain
simulation and can be used in the frequency scope of 0 similar to 100GHz.