How Baker Hughes is levering the Dell AI Factory with NVIDIA to Accelerate Centrifugal Compressor Simulations with Physics-Informed AI

Discover how Baker Hughes, Dell Technologies, and NVIDIA are pushing the boundaries of engineering with physics-informed AI. By accelerating CFD simulations up to 363,000 times faster while achieving 99.6% median accuracy, this breakthrough could transform how engineers design and optimize critical industrial equipment.
Oct. 23, 2026

What if engineers could reduce computational fluid dynamics (CFD) simulation times from 12 hours to just 0.1 seconds? Baker Hughes is exploring that possibility through groundbreaking research into physics-informed artificial intelligence (AI), supported by Dell Technologies and NVIDIA.

In this case study, discover how Baker Hughes is leveraging the Dell Pro Precision with GB300 and AI Physics to develop an AI surrogate model capable of predicting complex three-dimensional flow fields in centrifugal compressors. By overcoming traditional computing limitations, the research demonstrates how local AI infrastructure can accelerate engineering simulations while maintaining high levels of accuracy and protecting sensitive design data.

The results are compelling: approximately 363,000 times faster inference compared with conventional CFD simulations, 99.6% median accuracy in predicting 3D flow fields, and significant reductions in the engineering effort required to process complex models.

Download this case study to see how these advancements could transform how engineers design, test, and optimize critical industrial equipment as energy demand grows.