Session: 13-03 Heat Transfer Modeling Methods and Analysis

Submission Number: 178194

Radiation Heat Transfer in Turbomachinery: Accelerating Cyclic Symmetric Analysis of 2D and 3D Models 

Modern gas turbine development increasingly depends on advanced simulation to reduce physical testing and accelerate design cycles. As engines operate at extreme temperatures, accurate thermal modeling becomes critical for performance and reliability. Radiation heat transfer, particularly in combustors, cavities, and turbine blades, strongly influences component durability and cooling effectiveness.

Despite the availability of validated simulation tools, radiation modeling remains challenging due to complex geometries, surface thermo-optical property estimations, and trade-offs between accuracy and computational cost. Engineers often face a choice between oversimplified 2D models that compromise fidelity and highly detailed 3D models that exhibit longer simulation times and prolong design iterations. Failure to model radiation accurately can lead to underestimated heat loads, inadequate cooling strategies, and ultimately severe engine damage or reduced service life.

This paper presents radiation modeling approaches that balance fidelity and efficiency for turbomachinery applications. We start by presenting radiation modelling methods of full 3D geometries followed by simplifications including cyclic symmetry sectors with full and partial view factor computations, and then followed with further simplifications of geometries into 2D resulting in mixed 2D-3D hybrid models. Through the case studies, we compare and discuss the differences in modelling assumptions, accuracy and performance between the simplified approaches and full-expansion radiation methods, accelerated by GPU-based radiation computation. Lastly, practical guidelines are provided for selecting geometry representations, either 2D or 3D, based on accuracy, structural requirements, and computational resources.

By adopting these methods, engineers can reduce computational burden while maintaining confidence in thermal predictions. The proposed approaches streamline analysis workflows, support robust high-temperature designs, and alleviate common bottlenecks in radiation modeling—ultimately enabling faster iterations and improved engine performance.

 

Keywords: Whole Engine Modeling, Turbomachinery, Simulation, CAE, Multiphysics, Radiation, GPU, 2D-3D hybrid modelling, Heat Transfer

Presenting Author: Hussein Daou Maya HTT

Presenting Author Biography: Hussein Daou has a Master degree in Aerospace Engineering from McGill University. He has been working with Maya HTT for over 11 years, with a focus on Turbomachinery simulation applications. From the start he has been involved in development of the simulation products and has worked in close collaboration with numerous Turbomachinery companies in developing and evolving the software tools and capabilities. Currently, he's a Product Manager in the Thermal-Multiphysics team, focusing on Turbomachinery thermal solver applications.

Authors:

Alexandr Kuzmin Maya HTT
Hussein Daou Maya HTT
Audrey Collard-Daigneault Maya HTT
Florian Bonniol Maya HTT
Chris Barnes Rolls-Royce plc
Jeff Medema Rolls-Royce Corporation