Abstract: As the largest electricity distributor in Victoria, Australia, we deliver electricity to over 1.8 million residential households and commercial customers across Victoria. We are a data-driven energy organisation and exploit Big Data from over 1.8 million active smart meters in our Advanced Metering Infrastructure Smart Meter Network. We get 5-minute and 10-second voltage, current and power factor profile data of every smart meter in the network providing a time synced snapshot of the whole network.
Protecting the health and safety of our people and communities is our number one priority. Common safety emergencies communities encounter includes fallen powerlines, electric shocks and electrical fires. We are constantly looking at how we can continue to provide a high-quality and reliable electricity delivery service to a growing population, and then make it even better.
We have developed a variety of fault detection applications that aids in remotely identifying potential issues using data science deliverables, developed a distributed energy resource management system and outage forecasting application using machine learning techniques, and developed a real-time system automation using artificial intelligence – as some of the applications we have developed.
In our fault detection suite, we proactively detect neutral or active connection faults right up to the substation in Neutral Fault Detection application; detect potential high voltage network fuse issues in Brownout application and Candling Fuse Detection application; detect regulator issues in the high voltage network in High Voltage application; and detect network faults and low voltage network asset failures including FOLCB, FSD, junction box, isolator, POA and service wire in Network Fault Detection application.
In this talk, we will demonstrate how we process our Big Data, gain insights of our Smart Grid and Innovation applications, how we do detect faults from smart meter power quality data and meter channel alerts that resulted in reducing shock incidents in our network, how we automate monitoring of customer sites for immediate hazard detection, how we balance the grid from distributed energy resource management system, and how we automate fault detection analytics and jobs dispatch with no human intervention.
Bio: Thilaksha Silva has obtained a Doctor of Philosophy (PhD) in Statistics from Monash University, Australia. Thilaksha is skilled in data science for electricity distribution, statistics, time series forecasting, predictive modelling and big data analytics. She is adept at advanced data analytics with 10+ years of experience and has mastered in communicating the business value across the business and engaging audience with data science on a deeper level.