Siemens Gamesa Taps NVIDIA Digital Twin Platform for Scientific Computing to Accelerate Clean Energy Transition

Siemens Gamesa Renewable Strength is doing work with NVIDIA to generate physics-educated electronic twins of wind farms — groups of wind turbines utilized to create electric power.

The organization has 1000’s of turbines around the world that gentle up faculties, households, hospitals and factories with clear strength. In overall they deliver about 100 gigawatts of wind ability, plenty of to ability practically 87 million households on a yearly basis.

Digital representations of Siemens Gamesa’s wind farms will be developed utilizing NVIDIA Omniverse and Modulus, which jointly comprise NVIDIA’s digital twin system for scientific computing.

The system will aid Siemens Gamesa reach faster calculations to enhance wind farm layouts, which is predicted to lead to farms capable of generating up to 20 per cent far more electricity than preceding types.

With the world-wide level of once-a-year wind energy installations possible to quadruple among 2020 and 2025, it’s much more critical than at any time to increase the electricity made by each and every turbine.

The world-wide trillion-greenback renewable vitality business is turning to digital twins, like individuals of Siemens Gamesa’s wind farms — and a single of Earth by itself — to additional climate investigate and accelerate the clear energy changeover.

And the world’s immediate clean electricity technologies improvements necessarily mean that a dollar expended on wind and photo voltaic conversion devices today outcomes in 4x additional electrical energy than a dollar put in on the very same techniques a decade back. This has remarkable base-line implications for the changeover in the direction of a greener Earth.

With NVIDIA Modulus, an AI framework for acquiring physics-informed machine understanding versions, and Omniverse, a 3D structure collaboration and environment simulation platform, scientists can now simulate computational fluid dynamics up to four,000x more quickly than classic solutions — and view the simulations at substantial fidelity.

“The collaboration between Siemens Gamesa and NVIDIA has meant a good step ahead in accelerating the computational pace and the deployment velocity of our most current algorithms progress in such a elaborate subject as computational fluid dynamics,” said Sergio Dominguez,  onshore electronic portfolio manager at Siemens Gamesa.

Maximizing Wind Energy

Incorporating a turbine upcoming to yet another on a farm can adjust the wind stream and generate wake effects — that is, decreases in downstream wind speed — which direct to a reduction in the farm’s manufacturing of electric power.

Omniverse digital twins of wind farms will assist Siemens Gamesa to accurately simulate the impact that a turbine may well have on another when placed in shut proximity.

Applying NVIDIA Modulus and physics-ML models operating on GPUs, researchers can now operate computational fluid dynamics simulations orders of magnitude speedier than with classic procedures, like those centered on Reynolds-averaged Navier-Stokes equations or significant eddy simulations, which can acquire above a thirty day period to run, even on a 100-CPU cluster.

This up to four,000x speedup permits the swift and correct simulation of wake effects.

Examining and reducing prospective wake results in real time, whilst simultaneously optimizing wind farms for a range of other wind and climate scenarios, have to have hundreds or thousands of iterations and simulation runs, which ended up customarily prohibited by time constraints and expenditures.

NVIDIA Omniverse and Modulus enable precise simulations of the advanced interactions involving the turbines, utilizing superior-fidelity and substantial-resolution models that are based mostly on low-resolution inputs.

Discover additional about NVIDIA Omniverse and Modulus at GTC, operating as a result of March 24.

Check out NVIDIA founder and CEO Jensen Huang’s GTC keynote handle in replay:

Leave a comment

Your email address will not be published.


*