Applying a CMAC neural network to a photovoltaic system islanding detection

Abstract

This study proposed an islanding detection method for a photovoltaic (PV) power generation system based on a cerebellar model articulation controller (CMAC) neural network. First, the islanding phenomenon test data were used as training samples to train the CMAC neural network. Then, the photovoltaic power generation system was tested with the islanding phenomena. The CMAC only requires the adjustment of the weighting values of the memory addresses to be activated. Therefore, it features a reduced training time. Furthermore, because of the quantification of the input signals, the detection tolerance of the proposed method was enhanced. Finally, the islanding detection test results proved the feasibility of the proposed detection method for islanding phenomena.