부산대학교 도서관

Distributed generators are required to be integrated with sophisticated anti-islanding protection system (AIPS). However, most conventional AIPS fails to detect unintentional islanding event during perfectly matched loading conditions. To overcome this issue, in this article, a hybrid islanding detection algorithm has been proposed based on the rate of change of kinetic energy and root-mean-square value of absolute frequency deviation at the point of common coupling (PCC). The methodology presents a significant contribution prior to the state-of-the-art methods in terms of the disturbance injection mechanism through the q -axis current control loop of the grid-tied inverter. The method activates the disturbance signal injection mechanism only when an unintentional islanding event is suspected based on monitoring certain parameters at the PCC. This mechanism has been proposed to have zero nondetection zone, very fast operation, straightforward implementation, and significantly less impact on the power quality and stability of the power electronics system. The performance of the proposed method has been evaluated by conducting 200+ islanding and nonislanding case studies in simulation as well as an experimental platform. The proposed method is robust, accurate, and fast enough to detect an islanding event cleverly with a detection time as small as 120 ms.