Fast and accurate yield rate prediction of PCB embedded common-mode filter with artificial neural network

A fast and accurate yield rate prediction method of PCB embedded Common-Mode Filter (CMF) is proposed and implemented by applying artificial neural network (ANN), one of machine learning techniques. The proposed CMF is a mushroom-like structure which has four metal layers. The frequency responses of full-wave simulation and the outputs of trained ANN have good consistency. Furthermore, a substantial number of data sets with specific deviations can be applied with trained ANN to predict yield rate and find out sensitive parameters quickly.