Prof. Mehrdad Zangeneh to Deliver Plenary Lecture at ISROMAC19 in Tokyo on Physics-Enhanced Machine Learning for Turbomachinery

LONDON, UK - June 19, 2026: TOKYO, Japan / LONDON, UK - Advanced Design Technology Ltd. (ADT), the global pioneer in advanced aerodynamic and hydrodynamic 3D inverse design software, has announced that its Managing Director, Prof. Mehrdad Zangeneh, will deliver a prestigious Plenary Lecture at the 19th International Symposium on Transport Phenomena and Dynamics of Rotating Machinery (ISROMAC19), held from July 5-7, 2026, at Waseda University in Tokyo, Japan.

Prof. Zangeneh's keynote, titled "Overcoming the AI Data Bottleneck in Turbomachinery: Physics-Enhanced Machine Learning via 3D Inverse Design," will reveal a paradigm-shifting approach that resolves the most critical barrier to deploying artificial intelligence in industrial turbomachinery design.

Smashing the AI "Data Bottleneck" in CAE
While data-driven machine learning (ML) and neural network surrogates promise near-instantaneous performance evaluations, modern engineering faces a massive computational hurdle: the "data bottleneck." Traditional deep learning models require thousands of high-fidelity, computationally expensive Computational Fluid Dynamics (CFD) simulations simply to "learn" fundamental fluid dynamics and physical constraints.

This problem is severely worsened by conventional CAD-based geometric parameterization methods, which deform surfaces blindly. This brute-force approach generates thousands of non-physical, unmanufacturable shapes, wasting immense computing resources. Furthermore, "black-box" machine learning models often fail to generalize when predicting complex, highly non-linear flow structures, such as boundary layer separation or transonic shock waves.

The Solution: 3D Inverse Design as a Physics-Guaranteed Filter
Prof. Zangeneh's presentation will detail how to overcome this barrier by using 3D Inverse Design as a foundational, physics-guaranteed filter for machine learning.

Rather than forcing an AI to manipulate raw geometric coordinates and guess the aerodynamic outcomes, the inverse method flips the script: it utilizes specific aerodynamic design parameters, such as blade loading and circulation distributions, to directly derive the ideal 3D geometry. Because the mathematical framework ensures that every generated shape satisfies specified work and mass flow targets, the design space is restricted entirely to physically valid configurations from the very first step.

"By utilizing a 'physics-clean' data generation approach, engineers can achieve a 10x to 100x increase in data efficiency. This allows for the seamless training of high-accuracy machine learning surrogates using a fraction of the traditional data load.", Prof. Mehrdad Zangeneh.

Key Takeaways from the Upcoming Plenary Lecture:
• Bypassing Data Bottlenecks: Learn how restricting design spaces to aerodynamic parameters eliminates wasted, non-physical CAD iterations.

• Algorithmic Acceleration: Discover how advanced machine learning algorithms, such as the Reactive Response Surface (RRS), can be used to further supercharge optimization speeds.

• Real-World Validation: Explore practical multi-point optimization case studies spanning high-performance centrifugal compressors, axial fans, and Francis turbine rotors.

• Industrial ML Expert Systems: See how this physics-grounded approach can scale into high-fidelity expert systems capable of thriving under rigorous industrial conditions, demonstrated via a mixed flow pump stage example.

About the Lecturer
For the past 30 years, Prof. Mehrdad Zangeneh has been at the forefront of developing advanced turbomachinery design codes based on the 3D inverse design approach and automatic optimization. His prolific research has led to foundational industry breakthroughs in radial turbomachinery, including the suppression of secondary flows in radial and mixed flow impellers, alongside the mitigation of corner separation in vaned-diffusers.

In recognition of his contributions to mechanical and aerodynamic engineering, Prof. Zangeneh has been awarded Japan's Turbomachinery Society's Gold Medal, and the prestigious ASME Henry Worthington Medal.

80-86 Gray's Inn Road
London
WC1X 8NH
United Kingdom

Media Contact:
Advanced Design Technology, Ltd.
sales@adtechnology.com
www.adtechnology.com

About Advanced Design Technology, Ltd.
Advanced Design Technology (ADT) is a global leader in the development of advanced turbomachinery design methodologies, shortening development timescales while delivering breakthrough aerodynamic and hydrodynamic efficiency. Headquartered in London, UK, ADT's client base includes major global aerospace, automotive, power generation, and marine organizations.

About ISROMAC
The International Symposium on Transport Phenomena and Dynamics of Rotating Machinery (ISROMAC) is a premier global forum bringing together researchers, academics, and industrial engineers to discuss the latest advancements in aerothermodynamics, rotodynamic, and fluid-structure interactions of rotating systems. For more details on the plenary schedule, visit the official website at https://isromac2026.com/plenary-lectures/.

This release was published on openPR.