Health Monitoring in Smart Grid Using Big Data Perspective

Transition from traditional power grid into a smart grid involves collecting and processing a large amount of data from generation, transmission to consumer level. Data collected at each level differs considerably in terms of the parameters under observation and frequency of collection, satisfying the 5 Vs (volume, velocity, variety, value and veracity) requirements of big data. In this paper, we have explored a health monitoring system for smart grid that can be used to prevent failures and outages from taking place. These outages are typically a consequence of cascading failures and fault propagation due to the connected nature of the power grid. To mitigate these failures, wide area monitoring and control systems are required to identify the vulnerable and critical components of the system. The proposed proactive health monitoring system monitors various electrical and non-electrical parameters at the distribution level. These parameters are then used to classify the system status into one of the four predefined stability zones/states. The state information can be used by a central unit to determine whether more data is needed, and/or any control action needs to be taken. Since the initial processing uses a smaller subset of data, the processing can be done faster. If required, large subset, i.e. (Phasor Measurement Units) PMUs can be used for further processing. This novel idea can act as an intermediate system between already deployed Supervisory Control and Data Acquisition (SCADA) and PMUs.