Utilities test drive Oracle analytics to check EV demand
June 4, 2019
Utilities are test driving analytics from Californian software company Oracle to manage the influx of electric vehicles. The analytics should help utilities better plan for energy demand as cars move from petrol to the grid.
The use of electric vehicles (EVs) is growing at a record rate, with the International Energy Agency (IEA) predicting that the number of electric cars on the road will rise from 3.1 million in 2017 to 125 million in 2030. Enabling utilities to manage this new energy demand on the power grid intelligently, Oracle Utilities has unveiled what it says is a breakthrough in EV detection.
Tapping deep machine learning, Oracle Utilities Analytics Insights can identify the presence of an EV, show the time and frequency of charging, and disaggregate the energy being consumed by the vehicle with advanced metering infrastructure (AMI) data.
With this intelligence, utilities can reliably plan for the energy infusion needed to power EVs at scale and engage customers to charge at the times that are the least expensive for them and best for the health of the energy grid. The EV detection capabilities are being piloted by a number of utilities.
"With solar, wind and storage technologies now constituting 90 per cent of investment interest, the road is paved for deeper decarbonisation of the electricity sector," said Ben Kellison, director of grid research at Wood Mackenzie Power & Renewables. "The case for transport electrification has never been stronger and the rapid growth in investment interest from car manufacturers is a confirmation of the future consumer demand for EVs. Utilities are now faced with an increasingly clean and decentralised system and they need new data and analytic packages to support a new planning paradigm."
The influx of EVs could represent an average additional growth of one to four per cent in peak load on the grid over the next few decades, according to a report by McKinsey. While this may seem modest, the impact will be highly volatile and cause unpredictable spikes at the local sub-station and feeder levels in residential areas. This load is projected to reach as high as 30 per cent peak growth in certain urban areas that are hotspots for EV adoption.
While this transportation development represents an important step forward in reducing carbon emissions, most electricity grids were created long before EVs were a commercially viable consumer product. As transportation continues to evolve from petrol to the grid, utilities must plan for an uptick in energy demand that will vary dramatically by area.
"With almost every major auto manufacturer releasing new EV models in the coming years, the window of time for utilities to act is closing," said Dan Byrnes, SVP of product development at Oracle Utilities. "The intelligence our analytics provide is essential for utilities to make needed assessments on grid investments and, in tandem, work as trusted advisors to customers who may be in the dark as to how owning an EV is impacting their energy footprint and bill. From utility optimisation to proven customer engagement, only Oracle offers a complete package to manage the explosion of EVs."
The EV detection capabilities are powered by more than a decade of research and experience disaggregating household energy data from billions of data points collected from 60 million households across 100 utilities. Oracle's trained data models can be deployed for each specific household's usage to understand whether a customer has an EV, how they interact with their EV chargers, and where EVs are clustering on the distribution grid. As such, utilities will be able to plan for and manage the operational impact of EVs as a new distributed energy resource on the grid.
From a consumer perspective, charging an EV can increase a typical household's energy usage by 15 per cent or more and potentially double usage during peak demand times. With the offering, utilities will have the tools to roll-out intuitive, user-friendly EV adoption customer journeys and time-of-use plans to engage, educate and reward owners for charging during non-peak times. In the future, these same kinds of engagement programmes can be used for utilities to buy-back unused energy from their customers' EV batteries to help balance energy supply and demand in times of need.
"EVs will have an impact on every part of a utility's operations, from grid stability and regulatory affairs to customer billing and engagement," said Byrnes. "With Oracle, our customers have the tools and intelligence they need to make better decisions, maximise outcomes, and increase customer satisfaction every step of the journey."