AI Weekly: How the power grid can benefit from intelligent software

The U.S. electrical grid has long been burdened by aging infrastructure. Sixty percent of distribution lines have surpassed their 50-year life expectancy, according to Black and Veatch, while the Brattle Group anticipates $1.5 trillion to $2 trillion in spending by 2030 to modernize the grid and maintain reliability. The latest report from the American Society for Civil Engineers found that current grid investment trends will lead to funding gaps of $42 billion for transmission and $94 billion for distribution by 2025.

Neil Sahota, chief innovation officer at the University of California, Irvine, says intelligent software opens the door to the deployment of AI designed for power grid use cases. Utilities are already employing AI to address the windfalls and fluctuations in energy usage. Precise load forecasting ensures operations aren’t interrupted, thereby preventing blackouts and brownouts. And it can bolster the efficiency of utilities’ internal processes, leading to reduced prices and improved service.

“There are a lot of subtle clues that in aggregate show where and when a natural disaster can occur. To ‘see’ the clues, we need to process a lot of data across a broad spectrum of variables and look for subtle differences,” Sahota told VentureBeat via email. “This is difficult for people to do effectively but is in the wheelhouse of AI. Consider wildfires, where we are using climate information (including wind forecasts), drone surveillance, and satellite images to predict hot spots and how a fire may start and spread. AI can monitor all these millions of data points in real time and constantly generate prediction models.”

For example, startup Autogrid works with more than 50 utilities in 10 countries to deliver AI-informed power usage insights. Its platform makes 10 million predictions every 10 minutes and optimizes over 50 megawatts of power, which is enough to supply the average suburb. Flex, the company’s flagship product, predicts and controls tens of thousands of energy resources from millions of customers by ingesting, storing, and managing petabytes of data from trillions of endpoints. Using a combination of data science, machine learning, and network optimization algorithms, Flex models both physics and customer behavior, automatically anticipating and adjusting for supply and demand patterns.

O’Connell believes that efforts like X’s will face challenges, particularly on the distributed energy resource (DER) side of the equation. DER systems — small-scale power generation or storage technologies that provide an alternative to traditional power systems or enhance those systems — can be difficult to orchestrate because they might span solar panels, electric vehicle charging setups, and even smart thermostats. But if a digital transformation of the power grid succeeds, its long-term benefits could be significant, O’Connell says.

“Currently, when independent system operators want to add a new market participant type, it takes them a year to incorporate those changes. That’s legacy IT systems,” he said. “The IT systems that grid operators will need are going to have to get a serious upgrade from the ’90s technology that they use now.”

For AI coverage, send news tips to Kyle Wiggers — and be sure to subscribe to the AI Weekly newsletter and bookmark our AI channel, The Machine.

Thanks for reading,

Kyle Wiggers

AI Staff Writer