Tapestry and PJM Partner on AI for the Interconnection Queue

The country’s largest electricity market, PJM, has a problem: It’s facing a slew of retiring fossil fuel resources just as electricity demand is ramping, largely thanks to AI data centers. Meanwhile, PJM has a years-long waitlist full of wind and solar projects seeking permission to connect to the grid that are languishing in no small part due to its slow approval process.

Enter Tapestry, the so-called “moonshot for the electric grid,” as Page Crahan, Tapestry’s general manager, put it on a press call Wednesday. The initiative is a part of Google’s internal project incubator, known simply as X. Today, the tech company and the grid operator announced a partnership that will use artificial intelligence to develop a unified model of the grid’s electricity network, bringing in data from dozens of disparate tools into one simplified “Google Maps for electrons,” as Crahan put it.

The model will give grid operators and project developers the ability to toggle on and off different layers of grid information — a vast improvement over the technical boondoggle grid planners face today. As Crahan explained, they might use one software program that tracks where grid equipment is located, another that models power flow, another that measures the equipment’s thermal capacity, and yet another that runs an economic impact analysis. Then, Crahan said, “each of these software programs will generate a file which creates its own unique model of the grid. And every time a change is made to that one model, it needs to be applied to all of the other models in consideration.” Overall, the siloed nature of these different programs makes it a headache to keep information consistent and up to date across the entire system.

This convoluted process is partially at fault for PJM’s backlogged queue, which in recent years has seen a deluge in new interconnection requests, largely for renewable energy projects. Due to this system overwhelm, PJM put a pause on reviewing new applications in 2022, initially expected to last for two years. Now, it’s expected to lift early next year.

Aftab Khan, an executive vice president at PJM, said on the press call that the grid operator knows there’s still more work to be done. “Even though we've made significant progress with tools and automation to manage large numbers of projects in an interconnection cycle, it’s still, end-to-end, about a two-year process,” he said. To expedite this, Tapestry plans to deliver solutions that PJM can start rolling out this year. The two entities will work together to develop new processes “over the next several quarters “ and “perhaps even the next several years,” Crahan said.

Tapestry was formed in 2017 with the mission to “bring the grid out of the industrial age and into the age of intelligence.” In addition to creating a coordinated model of PJM, Tapestry is also developing an AI tool that automates much of the review process for grid interconnection applications, thereby helping to more efficiently validate the feasibility of proposed projects. It’s as simple as dropping a PDF into Tapestry’s AI analytics tool, which can then automatically check the data in the application against other reliable sources.

“By automating and improving the data verification process for things like land rights, equipment and grid impacts, we aim to reduce the burden on the PJM planning team and the energy developers,”Crahan said. She said this will help “reduce the time it takes to evaluate these projects so that capacity can come online faster.”

All of this work builds on previous projects and pilots that Tapestry has been running both domestically and abroad. For example, Tapestry has partnered with the U.S.-based utility and power company AES to develop a vision for the digital, AI-powered grid of the future. And in Chile, Tapestry worked with the national grid operator to deploy planning tools that enable speedy, long-term simulations, allowing operators to make informed decisions about energy needs decades into the future.

“We have been able to take a process that took the planners several days and turn it into a few hours,” Crahan said of the Tapestry’s work expediting grid simulations in Chile. Though she couldn’t cite specific targets for speeding up PJM’s grid interconnection process, “we're looking for significant order of magnitude improvement to support the PJM planners,” Crahan said.

Editor’s note: This story has been updated to clarify details concerning PJM's interconnection application review process.