AI can help increase wind generation, energy revenues, and cut maintenance costs.

AI is playing an increasingly bigger role in the management of wind energy, wind farms, and the maintenance of wind turbines through machine condition monitoring systems. Google (NASDAQ: GOOGL) predicts wind conditions and production for the next day for its 700 MW wind energy resources, gaining an edge in its power market operations. In Europe, the SmartWind project seeks to cut costs and optimize revenues, while a German company is using big data and AI for fault diagnosis of wind turbines. (AItrends.com)

Google’s DeepMind and wind forecasting

Google sources 700 MW of wind in the central US. Using a combination of weather and power data with machine learning, Google, and subsidiary DeepMind can more reliably predict wind production for the next day. As a result, they can get a better picture of the supply-demand balance and can therefore reduce operating costs.

“What we’ve been doing is working in partnership with the DeepMind team to use machine learning to take the weather data that are available publicly, actually forecast what we think the wind production will be the next day, and bid that wind into the day-ahead markets,” said  Michael Terrell, the head of energy market strategy at Google, at a recent seminar.

The process has paid off – revenue from the wind farms has jumped 20%.

In Europe: SmartWind

A consortium of four companies and the Ruhr-University Bochum in Germany are executing the SmartWind project in Europe.

The project will optimize the management of wind farms by using AI algorithms. A cloud-based platform will collect data in real-time from sensors and control systems. The algos will crunch this data and after diagnosis, will deliver recommendations on fault detection and O&M actions.

Wind farms can, therefore, reduce costs and optimize revenues.

Bruel & Kjaer Vibro (B&K Vibro) of Darmstadt, Germany

The company provides monitoring solutions for verifying the condition of machinery. Technically, these are known as machine condition monitoring systems. Its product range comprises vibration sensors (acceleration, velocity, and displacement), vibration monitors, handhelds, and rack-based plant-wide integrated monitoring solutions.

Over the past two decades, B&K Vibro has accumulated a vast amount of data on wind turbines “that includes fault data on almost every imaginable potential failure mode,” according to Mike Hastings, a senior application engineer with Bruel & Kjaer Vibro (B&K Vibro) of Darmstadt, Germany.

It includes both machine vibration and process data under all kinds of operating conditions. Moreover, the data also covers all kinds of wind turbine types and components, according to Hastings.

Condition monitoring systems (CMS) are now widely prevalent in wind turbines, particularly in remote and offshore installations.

Furthermore, AI and machine learning can be trained to perform various tasks for wind farms and their turbines on the data received from CMS.

These include fault detection optimization, automatic fault identification, and prognosis for failure.

Related Story:   Renewable Energy ETFs Soar On Possible Dem Election Sweep