Interleaved DC-DC boost converter in DC distribution fault detection method using Artificial Neural Networks

This paper proposes a fault detection method of the interleaved bi-directional DC-DC boost converter using Artificial Neural Networks (ANN). In the proposed method, when open-switch faults occur, fault detection is performed using the gating signal and the inductor current slope. This method can compensate for the delay time, and detect the fault fast within 2-sampling time in real-time. Through the ANN, fault detection is possible without additional circuits or complex algorithms, and training data is composed of integers, errors can be reduced. The proposed method is verified by PSIM simulation.