Tapestry, Google’s moonshot for the electric grid, is making its debut on an American transmission network, through a partnership announced today with PJM.
The pair will develop AI-powered tools to streamline and accelerate the process of connecting new generation to the grid — against a complicated backdrop of load growth, shifting federal stances on fossil fuels, and uncertainty around tariffs on power imported from Canada.
PJM, the largest grid operator in North America, is facing accelerating load growth due largely to new data center demand in the region. At the same time, the transmission operator expects more than 20% of its generation capacity — 40 gigawatts — to retire by 2030.
The massive backlog of projects in PJM’s interconnection queue is central to PJM’s challenges. The RTO currently receives thousands of interconnection requests annually, and has 200 GW of renewable power waiting in the queue.
And while Tapestry and Google are already developing tools for grid insights and planning in other pilots, the partnership with PJM marks the first direct application of Tapestry’s tools to manage interconnection queues. As Tapestry general manager Page Crahan put it: “This is the first time artificial intelligence is being used to manage the entire energy interconnection queue and process, as opposed to point solutions.”
The current interconnection process in PJM takes around two years from end-to-end, said Aftab Khan, PJM’s head of planning, operations, and security. It’s a process that involves building various models, developing cost estimates of needed transmission reinforcements, and developing schedules, he added.
But while there’s “significant potential” to make that process more efficient, Khan also said the RTO doesn’t have a target reduction goal for the partnership with Tapestry.
Crahan said Tapestry plans to deliver solutions that PJM can use starting in 2025.
Building “Google Maps for electrons”
One of Tapestry’s foundational goals, Crahan explained to Latitude Media last year, is to be “early in understanding the problem, but right on time with being able to deliver something people can use.”
Bringing its grid planning tools to the American grid has been a “top goal” for Tapestry, Crahan said, but breaking into the market has been challenging. That’s in part because access to data is more challenging in the U.S., she said.
Instead, Tapestry started its transmission-level work in Chile, partnering with the national grid operator there to build and pilot a planning tool for the country’s transmission grid. The focus of that pilot, which began in 2021, was specifically to help the country phase out coal ten years sooner than planned.
“We actually said out loud to Chile that our goal is to speed up the transmission planning by 100 times,” Crahan explained.
The pair are getting very close: According to Crahan, Chile’s grid operators can currently simulate their grid 86% faster, running 30 times the number of scenarios in the amount of time it used to take to run a single scenario. And in 2025, Chile is deploying the tool at full-scale for its annual transmission planning process.
PJM’s end goal for the partnership with Tapestry is a little broader than Chile’s. But some of the challenges are the same — Chile also faces grid congestion and high volumes of renewables in its queue, for example.
Tapestry’s work with PJM will build on those learnings, but will develop a new set of tools catered to the needs of an RTO, Crahan said.
One tool will “intelligently assess” PDFs, thereby automating the data verification process for interconnection requests. By reducing that burden on energy developers and PJM planners, Tapestry hopes to accelerate project approvals and get power projects onto the grid faster.
Another tool, Crahan said, will combine dozens of existing databases and tools PJM currently uses to evaluate requests, in order to create a “single, unified model of PJM’s network.”
