ADAPTIVE OVERCURRENT DISTRIBUTION PROTECTION SCHEME COORDINATION IN THE PRESENCE OF DISTRIBUTED GENERATION USING ARTIFICIAL NEURAL NETWORK

Recent advances in the development of distributed generation (DG) technology have resulted in increased DG penetration into distribution networks. The comparative cost advantage of its installation over grid expansion far outweighs its disadvantages. However, DG penetration introduces a lot of protection issues as a result of variation in the network short circuit current and bidirectional flow of current that, if not properly coordinated, causes maloperation of overcurrent relay protection scheme. This paper presents a novel adaptive protection scheme coordination system that uses artificial neural network (ANN) to coordinate the operations of the main and the backup relays. The proposed method determines the time settings multipliers (TMS) and the plug settings multipliers (PSM) based on the network's prevailing condition and communicates the new settings to microprocessor-based relays via optic fiber by using remote terminal unit to update the settings. The proposed scheme was applied to the IEEE 33-bus distribution network considering two different scenarios: with and without DG penetration. NEPLAN software was used to generate different fault currents required for the ANN training. Obtained simulation results show that the proposed adaptive protection scheme coordination significantly improves the protection scheme coordination with or without DG penetration.