Accidental disconnection of a live feeder section is a major concern accompanying large distributed solar photovoltaic (PV) generation integration. High localised penetration can alter power flows leading to anomalous occurrences. Load-inverter power balance during grid-side disturbances, for different load models, may cause unique situations that can trigger the interconnection point protective devices. One such phenomenon, identified as a potential cause of unintentional islanding on radial feeder models (a modified IEEE feeder in simulation and verified on a laboratory-hardware network), has been used in this work. A pre-emptive detection strategy has been implemented to identify such islanding initiators among other power system transients. Computational geometry concepts have been utilised to create an optimisation-derived feature extraction methodology for effective training of a classifier module realised in a Raspberry Pi microcomputer. This module predicts the class labels of test data points transmitted from simulations carried on a personal computer for the feeder model. The proposed pre-emptive islanding detection strategy can trigger an appropriate change in a PV inverter's operating mode before a feeder protective device is tripped by such island initiating anomalies. The online classification accuracy and speed indicate a possible integration of the proposed methodology and strategy with the inverter's control circuitry.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.