LAUSR

An intelligent load-shedding strategy was designed and embedded in the special protection system (SPS) to enhance the system stability for an offshore island with high penetration of photovoltaic (PV) systems. To prepare the training dataset for the artificial neural network (ANN), the transient stability analysis of the isolated power system was executed to determine the minimum amount of load to be interrupted to prevent the tripping of diesel generators for the emergency shutdown of PV systems. By selecting various combinations of PV penetration levels, total system load demand, and the frequency decay rate at the instant of PV system tripping as the input neurons of the ANN, the proper load-shedding scheme is derived and stored in the decision knowledge base of the SPS. When the intelligent energy management system detects the tripping of a PV system, the SPS will be triggered to determine the amount of loss to be disconnected and executes the corresponding load interruption. By applying the proposed ANN-based load-shedding scheme in SPS, the amount of customer loading to be interrupted has been reduced dramatically for the restoration of system stability after the emergency shutdown of high-penetration PV system.