Transparently mining data from a medium-voltage distribution network: a prognostic-diagnostic analysis

With the shift from the traditional electric grid to the smart grid paradigm, huge amounts of data are collected during system operations. Data analytics become of fundamental importance in power networks to enable predictive maintenance, to perform effective diagnosis, and to reduce related expenditures. The final goal is to improve the electric service efficiency and reliability to the benefit of both the citizens and the grid operators themselves. This paper considers a dataset collected over 6 years in a realworld medium-voltage distribution network by the Supervisory Control And Data Acquisition (SCADA) system. A transparent, exploratory, and exhaustive data-mining approach, based on association rule extraction, is applied to automatically identify correlations among SCADA events occurring before and after specific service interruptions, i.e., distribution network faults of interest. Therefore, both the prognostic and the diagnostic potentials of the dataset are investigated with respect to the occurrence of permanent service interruptions. Our results highlight a limited predictive capability of the available set of SCADA events, while they can be effectively exploited for diagnostic purposes.