Virtual designs make product development cheaper and faster

(30-03-2022) In his PhD, Leoluca Scurria investigated whether virtual modeling of machine parts can be further optimized and more specifically for industrial gearboxes.

The increasing demand for optimized and lightweight machines and machine parts as well as the continuous search of manufacturers for ways to quickly bring new products to the market is accelerating the integration of Computer Aided Engineering (CAE). CAE is the broad use of computer software in, for example, the development of machine parts.

Early in the design cycle, CAE tools are used to assess a large variety of possible design architectures. These analyses aim at pinpointing the most promising design variants that will then continue in the design cycle. At this stage of the design cycle, no precise information concerning the system is available, hence the models used for such analysis need to capture the global behaviour of the system while maintaining a low computational cost. This is of paramount importance to enable reasonably short design cycles.

"Towards the end of the design cycle, other more sophisticated CAE tools are deployed that refine the final product design in detail and ensure optimal performance. Such design tools allow physical prototypes to be partially replaced by virtual ones. In this way, development is cheaper and faster so that the product can reach the market faster," explains Leoluca.

"In my PhD, I investigated whether the modeling of industrial gearboxes could be further optimized, by employing high fidelity lubrication models. In order to enable the use of high fidelity EHL models in CAE tools, advance model order reduction techniques have been developed to drastically reduce the computational cost while retaining the same level of accuracy”, Leoluca concludes.

Read a more detailed summary or the entire PhD


PhD Title: High-Fidelity Modeling of Elastohydrodynamically Lubricated Contacts for System-Level Simulations


Contact: Leoluca Scurria, Dieter Fauconnier

Leoluca Scurria

Leoluca Scurria holds a bachelor's and a master's degree in Mechanical Engineering from the University of Pisa. He developed his master thesis as an intern in Siemens Digital Industries Software in Leuven on efficient numerical modelling of rolling element bearings in lubricated conditions.

Successively, he was awarded a Baekeland mandate for a project between UGent, Siemens Digital Industries Software and KU Leuven. This project aims at advancing the state of the art in efficient lubricated contact modelling. The main research question in this project is: is it possible to use high fidelity numerical models of lubricated contacts in system-level simulations of mechanical transmissions? Leoluca's research showed that this is possible using the developed modelling strategy and that the level of accuracy for the target cases is similar to the accuracy of the more complex Computational Fluid Dynamics models.

These advances in the state of the art resulted in two journal publications. The first one, concerning the accuracy of the proposed method with respect to computational fluid dynamics, was published in the Journal of Tribology. The second one proposing a novel model order reduction technique to improve computational efficiency was published in the journal Computer Methods in Applied Mechanics and Engineering.


Editor: Jeroen Ongenae - Final editing: Ilse Vercruysse - Illustrator: Roger Van Hecke