Skip to main content

Manufacturing

Digital Twin: Wind Farm

Monitor three wind turbines, investigate an incident, and test what-if responses in 3D. Follow four steps from farm output and equipment signals to a possible cause and a simulated outcome.

Demo data is synthetic. Investigation histories are illustrative; simulated outcomes are estimates, not validated forecasts.

How it's builtAWS IoT TwinMaker, AWS IoT SiteWise, AWS Lambda, Amazon SageMaker, React, TypeScript, Python

AWS IoT SiteWise stores synthetic turbine telemetry, and AWS IoT TwinMaker defines the scene. React and React Three Fiber render the 3D interface. AWS Lambda runs case analysis and isolated what-if experiments; Amazon SageMaker scores deviations from synthetic normal operation.

Features

The real turbine explorer in Cutaway mode with the exposed drivetrain and original inspection controls

See the farm. Inspect a turbine.

Explore three turbines across two zones, switch between exterior, cutaway, and exploded views, and select a measurement to focus its component.

The real incident history showing gearbox oil temperature rising over 24 illustrative hours to 102 degrees Celsius

Trace the incident. Follow the signals.

Investigate four incident types affecting one or two turbines, and inspect wind, power, temperature, and vibration on a 24-hour history.

The actual Windfarm comparison of the original gearbox cooling incident and the calculated outcome after an assumed cooling repair

Test a response. Compare the outcome.

Adjust wind, yaw, and operating controls or assume a repair, then compare the original case with the simulated outcome.