Enhancing the reliability of protection scheme for PV integrated microgrid by discriminating between array faults and symmetrical line faults using sparse auto encoder

• DOI: 10.1049/iet-rpg.2018.5627
• Publication Date: 2018-10-31T02:38:23Z
• Authors: Murli Manohar; Ebha Koley; Subhojit Ghosh
• Keywords: Microgrid; Line (geometry)

The ever increasing power demand and stress on reducing carbon footprint have paved the way forwidespread use of photovoltaic (PV) integrated microgrid. However, thedevelopment a reliable protection scheme for PV microgrid ischallenging because similar voltage‐current profile array faultsand symmetrical line faults. Conventional schemes based onpre‐defined threshold setting are not able to distinguish between andsymmetrical faults, hence fail provide separate controlling actions forthe two cases. In this regard, sparse autoencoder(SAE) deep neural network has been proposed discriminate arrayfaults faults in addition perform mode detection, faultdetection, classification section identification. voltage‐currentsignals retrieved from relaying buses converted into grey‐scale images andfurther fed as input SAE unsupervised feature learning. Theperformance evaluated through reliabilityanalysis compared with artificial network, support vector machine anddecision tree techniques under both islanding grid‐connected ofthe also validated field applications byperforming real‐time simulations OPAL‐RT digital simulator.