According to MIT Technology Review, California’s nationwide power grid operator is on track to be the first in North America to employ artificial intelligence to maintain outages. &nbsp,

” We wanted to update our network management systems. This completely fits that,” says Gopakumar Gopinathan, a top consultant on energy system technology at the California Independent System Operator, also known as the CAISO and pronounced KAI-so. ” AI is now transforming various companies. However, there aren’t many instances in which our business has used it.”

On July 15 at the DTECH Midwest power market conference in Minneapolis, CAISO is scheduled to make an announcement about a partnership to work a pilot program using the new AI technology Genie from the energy-services provider OATI. The software has the ability to freely decide decisions about important grid functions, such as switching from armed traffic officers to sensor-equipped stoplights. It uses conceptual AI to evaluate and perform real-time analyses for grid operators. &nbsp,

However, CAISO’s role in managing the government’s electrical system is remarkably analog, despite the fact that it does offer electrons to cutting-edge Silicon Valley businesses and laboratories. &nbsp,

Now, CAISO engineers look through failure reports for keywords related to maintenance that is planned or in progress, read through the notes, and then fill each piece into the grid software system to calculate how a broken line or transformer may impact the power supply.

When you multiply that number by 200 or 300, according to Abhimanyu Thakur, OATI’s vice president of platforms, representation, and analytics, “even if it takes less than a second to scan one on regular,” Thakur says. Therefore, various agencies are using their own distinct keywords. Now that we have consolidated everything into a single lexicon of keywords, AI can perform this record and create a report proactive.

Gopinathan suggests that the operator might regard automating more features on the grid if CAISO discovers that Genie produces reliable, more effective files analyses for managing disruptions. He claims that after a few tests rounds, we will know whether or not it is appropriate to call it effective. &nbsp,

Regardless of the outcome, the test represents a major change. According to Richard Doying, who spent more than 20 years as a leading executive at the Midcontinent Independent System Operator, the grid operator for an area that includes 15 claims from the upper Midwest to Louisiana, the majority of network users are using the same techniques that resources have used” for years.” &nbsp,

Doying, a vice president at the consultancy Grid Strategies, says,” These organizations are carved up for people who are working on very specific, specialized tasks that they’ve developed over the years. Grid operators are only benefited by the fact that some of these new AI tools can use data from various organizational areas to perform more sophisticated analysis.

A report from the Department of Energy last year demonstrated that AI had the potential to accelerate studies on grid capacity and transmission, improve weather forecasting to help determine the amount of energy wind and solar plants can produce at a given time, and improve planning for electric-vehicle charging networks. According to another report from the Energy Department’s Loan Programs Office, adding more “advanced” technology like sensors to various pieces of equipment will produce data that will enable AI to do much more over time. &nbsp,

The PJM Interconnection, the largest grid system in the country, which spans 13 states along the densely populated mid-Atlantic and Eastern Seaboard, signed a deal with Google to use its Tapestry software to improve regional planning and speed up grid connections for new power generators in April. &nbsp,

According to a source familiar with the plans who requested anonymity because they were not authorized to speak publicly, ERCOT, the Texas grid system, is considering using technology similar to what CAISO is currently planning to use. ERCOT did not respond to a request for comment. &nbsp,

Australia provides an illustration of what the future might look like. AI software developed in New South Wales, where grid sensors and smart technology are more widely used, is now predicting the state’s rooftop solar units ‘ production and flow of electricity and automatically adjusts the amount of electricity that can enter the grid from those units. &nbsp,

The electricity demands of AI data centers have dominated the discussion so far ( check out MIT Technology Review‘s Power Hungry series for more on this ).

Charles Hua, a coauthor of one of last year’s Energy Department reports and current executive director of Power Lines, a nonprofit that advocates for improving the affordability and reliability of US grids, says,” We’ve been talking a lot about what the grid can do for AI but not nearly as much about what AI can do for the grid.” ” Generally speaking, there is a huge opportunity for grid operators, regulators, and other stakeholders in the utility regulatory system to use AI effectively and make use of it to create a more resilient, modernized, and strengthened grid.”

Gopinathan claims that he is still holding back somewhat optimistic for the time being. &nbsp,

He says,” I don’t want to overhype it.” &nbsp,

He continues, “it’s a first step for bigger automation.”

This is more limited to our outage management system at this time. Genie isn’t talking to our other systems yet, he claims. However, I see a world where AI agents can accomplish a lot more.