In a Changing World, Energy Providers Need AI
image credit: ID 159515752 © Everythingpossible | Dreamstime.com
- Dec 2, 2020 5:55 pm GMTNov 25, 2020 3:40 pm GMT
- 844 views
This item is part of the Special Issue - 2020-12 - Data Analytics & Intelligence, click here for more
Digital analytics and machine learning are key to beating climate change and COVID-19
It’s no secret that we’re currently living through a period of historic disruption. The COVID-19 pandemic has transformed our economy, shuttering schools and businesses and shifting many workers into remote work. Simultaneously, the climate crisis continues to spiral out of control, with wildfires, flooding, storms, and heatwaves both changing the way we use energy, and creating enormous logistical problems for energy providers.
To overcome those challenges, energy companies need to understand the unprecedented short-term and long-term changes that are now impacting power demand. In other words, we need to know both what’s going to happen to energy demand in the next hour, next day, or next week, and what’s going to happen six months or several years from now, based on longer-term fluctuations in weather, economic factors, electrification technologies, and related issues.
Understanding how COVID changed energy usage
Fortunately, we now have the tools we need to understand our changing world and generate accurate demand forecasts. By applying the latest innovations in artificial intelligence to the enormous amounts of data being generated by smart meters and other smart-grid infrastructure, we can spot new trends and changing usage patterns more accurately than ever, and in real time.
At Innowatts, for instance, we recently used AI-generated insights gleaned from more than 40 million meters with granular consumption data to reveal the impact of COVID-19 lockdowns on different sectors of the U.S. economy. Some of the findings were much as you’d expect: with schools closed, educational institutions’ energy usage fell 40%. Others were less intuitively obvious: liquor stores, pizza parlors, and storage units all saw energy usage increases.
We also found that residential consumption was between 6% and 9% higher than normal, and up to 15% higher during working hours. Load patterns also changed: mealtimes became optional, with lunch and dinner peaks evaporating as home-bound families cooked at whatever time suited them best.
Such research shows the ability of AI analytics to help energy providers understand shifts in energy usage as they’re happening, enabling them to better and more accurately plan for the future.
Providing better and more accurate short-term forecasting
AI-enabled analytics can be used to understand more immediate demand curves, too. Energy companies typically use historical data to understand how factors such as weather and economic trends will drive short term energy use. But in a changing world, amidst increased weather volatility and economic transformation, historical data alone won’t work.
By tapping into smart meters for near real time usage information, AI tools can generate markedly more accurate day-ahead and week-ahead demand forecasts, giving energy providers the tools they need to more easily match production to demand and keep costs low for everyone.
That’s important, not just because it helps providers to keep rates low and profits high, but because it gives them the certainty they need to blend higher proportions of fluctuating renewable generation. Energy providers that capitalize on customer-centric intelligence-led AI transformation will be able to put customers first not only by anticipating changes but by driving them, and proactively managing costs, revenues, and customer satisfaction.
Decarbonizing the grid
The need for accurate, AI, and smart meter driven energy demand analytics and forecasting will only increase in coming months. With President-elect Joe Biden pushing for a zero-carbon electrical grid, we’ll see new decarbonization strategies, including grid-connected electric vehicles and other distributed storage technologies, proliferating across the U.S. in coming years.
To promote new, cleaner energy technologies, energy providers will need to ensure that they have a clear and granular understanding of short-term and long-term changes in demand curves, enabling them to make smart, strategic, and dynamic decisions about their evolving generation mix. This is also important as providers seek to mitigate increased risk from extreme weather events.
Utilizing those intermittent (uncontrollable) resources efficiently at scale, while keeping the lights on and costs low, will require the ability to quantify demand that can be shifted up or down (i.e. controlled). The AI tools we’ve developed to better understand changing demand drivers will play a critical role in helping us to build the sustainable, resilient energy system of the future.