Modeling of EMI Filter by Means of Recurrent Neural Network with Encoding Layer

In order to solve the electromagnetic interference (EMI) problems caused by high-speed switching of power electronic devices, EMI filters are widely used. However, the design of EMI filter is a time consuming and costly process. This letter proposes a new modeling method of EMI filter by means of recurrent neural network (RNN) with encoding layer. In this method, mutual coupling between components and topology of EMI filter are taken into account. The accuracy of the method is verified through comparing simulated and measured results of insertion loss of EMI filter. Compared with the traditional modeling method and other neural network-based methods, the proposed modeling method is more accurate and efficient under conditions of relatively small experimental data size.