Grid operator SPP is in discussions with Amazon Web Services to use artificial intelligence to accelerate interconnection research for utility-scale solar. The NextEra executive is taking the idea to other network operators.

Artificial intelligence, or machine learning, can improve the speed and accuracy of modeling for interconnection studies for large-scale renewable energy projects, Xin Wang, global network modernization leader for Amazon Web Services (AWS) Energy and Utilities, said during a panel discussion by trade group ACORE.

One type of interconnection study uses a model to estimate how a new solar generating system will affect the flow of energy on the grid. The model predicts energy flow “but doesn’t solve,” Wang said, meaning it doesn’t provide a solution. “You have to figure out where the problems are, and that takes years of engineering experience. We have a limited number of people who know how to do this.”

Machine learning can help, Wang said, if it learns enough cases to make “some suggestions or recommendations, like where energy flow is a problem.”

Wang spoke of “hundreds of scenarios” that grid operator SPP manages when it conducts interconnection studies, where engineers have to “see where the disturbances are, where the overloads are.” Then “you need to spread the costs of clearing those congestions back to the developers. And this is another rather complicated process. I think all of this can be solved with AI and machine learning if we make the right model.”

Grid operator SPP, which serves the central US, is in discussions with AWS about both artificial intelligence and more basic improvements such as automating data entry and validation. SPP has “some of the best engineers, but we’re not making the best use of their talents,” said panelist David Kelly, SPP’s director of seaming and tariff services.

Instead of engineers spending time copying data from applications, Kelly said, “we should use the talents of our engineers to make better recommendations about whether we need this new 345 kV line from point A to point B, or whether we need a new transformer substation. somewhere, and even that could come with recommendations from artificial intelligence.”

SPP has been working on solutions with AWS for the past year, Kelly said, and is working to finalize a formal agreement.

AI computing applications can serve other regions in the same way, said panelist Matt Pawlowski, head of business management and regulation at NextEra Energy Resources, which originally brought together AWS and SPP executives.

Pawlowski praised network operator MISO for having “some automated management models,” saying “these are models that other RTOs can pursue, that can be considered plug and play.” And this is what we strive for. We’ve had conversations with other regions, they haven’t been as advanced as SPP, but we’re getting there.”

In response to a question comparing the speed of SPP in completing studies of the relationship with fast pace Texas grid operator ERCOT, Kelly said the main advantage of ERCOT is that “cost sharing is defined” and noted that SPP has moved quickly to add 33 GW of wind power and now plans for the next 5 to 10 years.

Pawlowski agreed with ERCOT’s preference for cost sharing and added that ERCOT did not address the systems studies. He said one shortcoming of ERCOT is that it follows the British plug and play has higher congestion on transmission lines which leads to coagulation some renewable generation.

This content is copyrighted and may not be reused. If you would like to collaborate with us and want to reuse some of our content, please contact: