Machine Learning Model for a Trace Referenced to Meshed Ground Planes
Jiyue Zhu, Xiaoyan Xiong, Gang Kang, Karthikeyan Mahadevan · 2024
Transmission lines (TL) with meshed ground planes have been widely used in Flexible Printed Circuit Boards (FPCB) because of its physical flexibility and weight. Unfortunately, the introduction of the meshed ground plane increases the complexity of modeling for the signal integrity of TL, which is beyond the capability of TL theories and 2D numerical methods. Although 3D full wave methods can characterize the electric properties accurately, it costs time and computation resources heavily. In this paper, we propose a machine learning model to calculate the inductance and capacitance per unit length (PUL) of TL with meshed ground planes. In this model, multiple physical parameters which can affect the characteristic impedance are taken into account. Simulation samples are obtained using 3D full wave simulations as training and testing data. The testing results show that predictions of the machine learning model are in good agreement with the results of 3D full wave simulations.