How A.I. Batteries and Time-of-Use Rates Can Help Achieve Cost Savings For Owners / Managers
New Research from Prescriptive Data Shows That TOUs Can Incentivize Owners to Reduce Building Operating Costs and Help the Environment
NEW YORK (September 23, 2019) –Although the commercial building sector remains one of the largest contributors of all electricity usage in the U.S., new research shows that landlords who take advantage of time-of-use (TOU) electricity rates and battery electricity storage systems can reduce their peak monthly energy usage by up to 26%.
The report, co-authored by Prescriptive Data’s Head of Data Science Ali Mehmani and Columbia University research scientist Christoph J. Meinrenken, found that TOUs can incentivize owner/managers to install batteries in their buildings, which can play an essential role in reducing peak energy demands and thus increasing the efficiency and reliability of the power grid.
“With climate change on everyone’s minds, there are numerous ways that property owners can save money, reduce energy consumption, and help the environment,” Mehamani said. “Through our research, we found that batteries themselves have the potential to change the commercial real estate and building landscape, specifically within cost reduction on peak demand, as well as the implementation of tenant comfort through measuring interior temperature, and using machine learning to correlate those two things to optimize cost savings.”
The report tracked commercial properties across New York City using Con Edison’s TOU rate framework. Under the program, customers can pay less for electricity during off-peak periods, and more during peak times (8 a.m. – 6 p.m.) and summer super-peak times when the grid is busier and energy is more expensive. The study found that TOUs, together with the integration of batteries, were effective at shifting electric demand to off-peak hours. Batteries can be charged during off-peak times, and the stored energy can be used to power buildings at peak times when electricity is costlier.
Once the battery has been installed, owners can use an AI-based algorithm to optimize the usage of the battery during the day. Machine learning technologies can detect building data – such as space temperature, historical weather and electricity load – in order to automatically schedule the best times to purchase electricity.
Among the highlights from the report:
Up to 20% of the total installed electricity generation capacity in the United States is dedicated to meeting peak demands.
The building sector contributes up to 75% of all electricity usage and is a disproportionally large contributor to peak energy demand.
A 26% peak reduction means a dramatic improvement of grid stability and also can help reduce air pollution from often dirty peak generation capacity. Payback time is approximately five years.
The full report is available from the November 2019 edition of Applied Energy here.
