Digital Twin detects incorrect control system setting in wind turbine using remote monitoring

Analytics tool reveals that operator’s power curve performance was reduced by 2% - stopping USD 80,000 loss over the turbine’s lifetime

• Cost reduction
• Vast performance improvements
• Detection of performance errors that most analyses would miss

The challenge
To improve the performance and operating life of their wind farms, wind farm owners are continuously looking for solutions that allow them to gain the best insights into the health of each individual turbine on an ongoing basis.

While sensors and data monitoring systems indicate the condition of an asset, they can miss subtle changes in performance. Swift notification of a problem is needed to prevent small losses adding up to substantial sums. Conventional analysis is laborious and time consuming – skilled analysts are too often tied up performing routine procedures, when they should be looking for solutions to problems.

DNV’s solution
In response to the challenge of delivering real-time insights into individual turbine performance, DNV developed WindGEMINI, the digital twin analytics tool for wind turbines. The system’s algorithms produce a range of insights, from predictive maintenance to forecast of long-term energy production, including power curve analysis, component failures, and even remaining life, enabling turbine owners to improve the operation of their wind farm.

The system automatically analyses wind farm data and provides data-driven, actionable insights focused on critical aspects of the turbine that can greatly improve its performance – all this analysis is being done remotely and the results are directly transmitted to the wind farm operator to decide on the next course of action. DNV’s team of global wind experts is available to provide further diagnosis and advice.

Benefits
A wind farm operator in the U.S. implemented a control system upgrade across a project. All the turbines were quickly up and running again. However, WindGEMINI’s Pattern of Production module identified a slight decrease in the production from one of them. The power curve performance was reduced by just 2% - a change so small that most analyses would miss it, but that would have cost USD 80,000 over the turbine’s lifetime. Further investigation revealed that the turbine had been left with an incorrect pitch setting. The error was quickly resolved, and the turbine returned to full energy production.