Emerson, Vayu partner on machine learning for wind energy farm optimization

May 08, 2019

Emerson has formed an alliance with Vayu to provide automation technology solutions for wind farm energy optimization in the Americas, Caribbean and Europe. The three-year collaboration combines the advanced power applications and networking capabilities of Emerson’s Ovation automation platform with Vayu’s cloud-computing wind energy optimization technology. According to Vayu, it has identified more than $500 million in revenue opportunities from just a fraction of the approximately 450 wind farms in the United States using its technology.

"Combining the respective strengths of Emerson and Vayu creates a first-of-its-kind, intelligent solution for wind farm optimization," said Bob Yeager, president of Emerson’s power and water business. "This initiative will help wind energy producers maximize their aggregate power output, achieve their financial objectives and deliver more clean power to their communities."

According to the World Wind Energy Association, the wind power market continues to grow with 53.9 gigawatts added in 2018, bringing the overall capacity of all wind turbines installed worldwide to 600 gigawatts. The United States—the second-largest wind power market—added 7.6 gigawatts of capacity last year.

According to Emerson, in traditional wind farms, each turbine is individually optimized. Each creates its own turbulence, or wake, which prevents downwind turbines from receiving full energy from the wind stream, significantly reducing wind energy production. Vayu estimates two-thirds of U.S. wind farms experience reduced capacity due to wind wake. Emerson and Vayu’s solution leverages key data and machine learning to enable turbines to work cooperatively, adjusting the side-to-side movement of each turbine based on wind speed, wind direction and other parameters to extract the optimum energy from wind.

Emerson’s Ovation compact controller will communicate data from each turbine’s wind sensor to the Vayu system every 1–2 seconds. This typically takes 10 minutes when using traditional equipment.