Case Study

Machine Learning Aerodynamic Probes

At Cambridge Aerothermal we have been developing a pool of expertise around Machine Learning, and the benefits it can bring to measurement precision.

A series of Machine Learning aerodynamic probes (5-hole, 7-hole and 9-hole) have been developed in partnership with the Whittle laboratory at the University of Cambridge.

[Ref: GT2019-91428 Proceedings of ASME Turbo Expo 2019]

Benefiting from the extensive aerodynamic measurement experience at the Whittle lab, the new probe designs and data reduction algorithms are capable of increasing measurement resolution, reducing measurement uncertainty, and significantly increasing probe incidence tolerance when compared to probes with traditional calibration algorithms.

Machine learning aerodynamic probes can be supplied on request.