Physics-Enhanced Machine Learning for Full Stage Industrial Compressor Optimization
Thursday, October 22, 2026 | 4:00 PM (BST).
In this webinar, you will discover
• Parametric Stage Definition: How to rapidly define and optimize the shape of an industrial radial compressor stage, and how to give the system enough design freedom with just a handful of controlling input parameters.
• Unified Aerodynamic Optimization:How a Physics-Enhanced Machine Learning system goes about creating the most efficient 3D blade shapes in the given meanline outline, and matches diffuser and volute size to the flow coming from the rotor.
• Rapid Local Deployment:How to build and operate these Intelligent, physics-bound systems to run on your own local hardware in a matter of hours.
Complete the short form to register for the webinar.
.
Physics-Enhanced Machine Learning for Full Stage Industrial Compressor Optimization
ADT’s Physics-Enhanced Machine Learning system can be tasked with creating full stage turbomachinery, including rotors, diffusers and volutes.
By designing these components as a fully matched system we can exploit and extract the maximum flow performance and the highest efficiency.
We consider surge and choke margin so that map width becomes an optimisation objective, along with other multi-point performance considerations.
This webinar will demonstrate the workflow to set up and run this study - showing how easy it is to create a Physics-Enhanced Machine Learning system for an application that has proven stubbornly resistant to performance gains when using direct design techniques and separate component design considerations.
Key Takeaways
TURBOdesign suite has all the tools required to design full radial compressor stages, from scratch, with acute predictions of multi-point, multi-objective performance
3D Inverse Design is the technology that unlocks Physics-Enhanced Machine Learning
Training datasets of less than 60 points can be enough to find optimal designs, if those points contain the physics-linked parameters that describe the flow phenomena
Who is it for?
The event addresses all engineers, developers or researchers dealing with Turbomachinery Design.
TURBOdesign Suite Toolkits
3D Inverse design is used as the geometry engine to produce physics-enhanced design candidates
Variation in impeller geometry results in changes to diffuser inlet flow, so the diffuser geometry is updated to match accordingly
A range of possible diffuser widths and heights is allowed in order to maximise the stage efficiency
Meet the Speakers
Lorenzo Bossi
Chief Operating Officer
Rich Evans
Applications Engineer
Complete the form to have the recording sent straight to your inbox. Click Here
/PEML-compressor-optimization.jpg?width=1200&length=1200&name=PEML-compressor-optimization.jpg)
/tds.jpg?width=1200&length=1200&name=tds.jpg)
/cfd.jpg?width=1200&length=1200&name=cfd.jpg)