A Smart IEC 61850 Merging Unit for Impending Fault Detection in Transformers

This paper proposes and experimentally validates the functionality of a smart IEC 61850 merging unit (MU) that supports self-healing and asset management functions of future power grids. The proposed MU can operate in a standalone or as an integrated element within a primary substation. The MU communicates with a supervisory control and data acquisition (SCADA) system over Ethernet and WiFi-5 GHz links. A dynamic wavelet-based windowing technique is implemented in the proposed MU to process signals and report limited situation awareness (SA) features. The SA features serve two purposes. First, they can be used by asset management functions to monitor and diagnose the equipment heath condition. Second, they can be employed by self-healing functions in order to detect and anticipate early stages of impending faults masked by high noise. The MU capability to acquire real-time sampling or estimated-time sampling is examined and the impact of wireless sensor network data transfer latency between the proposed MU and an SCADA system is also investigated. Extensive laboratory experiments show promising results for the proposed IEC 61850 MU as a smart tool that can monitor, control, protect, and also initiate corrective actions.