Can artificial intelligence assist connect more photovoltaic and wind farms to the power grid?

The United States Department of Energy (DOE) believes that artificial intelligence can speed up the process of connecting recent energy projects to the power grid.

We plan to operate artificial intelligence to improve the interconnection application process, which is currently painfully ponderous. The delay is holding up efforts to launch recent solar and wind farms. Pressure is mounting to catch up in the face of rising electricity demand – ironically, in huge part because of the artificial intelligence boom.

Currently, it takes up to seven years for a recent electricity generation project to connect to the U.S. grid, and the wait is getting longer. As a result, there is a massive backlog of 2,600 gigawatts of recent energy generation and storage projects waiting to be interconnected. According to the Department of Energy, this is roughly twice as much as the United States currently has in terms of electricity generation.

The ponderous process of interconnection is also a legacy of an energy system designed around fossil fuels, when developers would have relatively few huge power plants in the pipeline. Power grids that operate more renewable energy tend to be less centralized, with electricity coming from a more distributed network of smaller solar, wind and battery projects. This means more applications to review.

Generally, these are recent photovoltaic and land farms cheaper sources of electricity than coal or gas today and do not cause air pollution or contribute to climate change. These benefits can be seen reflected in the type of energy infrastructure being built in the US. Above in 94 minutesT recent capacity waiting to be connected to the grid is emission-free energy, mainly solar energy, wind energy and batteries.

The DOE suggests that utilities and grid operators could more quickly implement interconnection requests for these projects if they could operate artificial intelligence. The current manual application review process is said to be labor-intensive, and the problem is compounded by the fact that applications submitted are often incomplete. The Department of Energy says communicating with developers to fill in missing information to correct applications can lead to significant delays.

Through AI4IX, the DOE wants to leverage existing artificial intelligence algorithms to quickly detect flawed applications and notify their authors. For example, tools can operate AI software trained in right applications. DOE is accepting proposals for the first round of AI4IX funding through January 10, 2025, and expects to announce awardees in winter 2025.

What’s more, Lee Zeldin, Trump’s pick to head the Environmental Protection Agency, says he wants to “make America the artificial intelligence capital of the world.” The pledge reflects the deregulatory agenda of the incoming Trump administration, particularly when it comes to any rules that could hinder the development of energy-hungry AI data centers.

Electricity demand in data centers could escalate by 160 percent by 2030 with the development of artificial intelligence, according to research by Goldman Sachs. Thus, increasing online electricity generation to meet this demand may continue to be a DOE priority.