Below, we present an overview of selected research projects conducted by our group.
Finite Element Analysis of Microstructural Evolution during Dry Machining of Additively Manufactured Ti6Al4V Alloy
This project develops and validates a three-dimensional finite element modelling framework to simulate the dry machining of Electron Beam Melted (EBM) Ti6Al4V, a titanium alloy widely used in additive manufacturing. The model predicts key thermo-mechanical loads and microstructural responses—including cutting forces, temperature distribution, strain, nanohardness changes, and α-phase lamellae thickness—under varying machining conditions such as cutting speed, feed rate, and tool geometry. A custom user subroutine bridges machining parameters with resulting microstructural evolution, enabling systematic analysis of how tool geometry and cutting angles influence machining performance and surface integrity. The validated simulation provides insights into optimizing machining strategies for additively manufactured Ti6Al4V components, improving process efficiency and final part quality.
A holistic analysis of laser powder bed fusion process parameters for Inconel 625 superalloy: microstructural features and mechanical performance
This project provides a comprehensive study of how key process parameters in laser powder bed fusion (L-PBF)—including laser power, scanning speed, layer thickness, and build direction—jointly influence the microstructure and mechanical performance of the nickel-based superalloy Inconel 625. Rather than examining parameters in isolation, the research systematically explores their combined effects to uncover critical relationships between processing conditions, melt pool geometry, texture development, and resulting properties. Optimized parameter combinations produced Inconel 625 specimens with a notably high combination of tensile strength and ductility, while detailed fractographic analysis revealed transitions in fracture behaviour and strong crystallographic texture linked to enhanced performance. The findings offer a practical framework for tailoring L-PBF settings to meet demanding structural requirements in aerospace, automotive, and biomedical applications.
Neural network-based prediction of plastic strain-hardening in metal additive manufacturing
This research develops an artificial neural network (ANN)–based modelling approach to predict the plastic strain-hardening behaviour of metals produced by additive manufacturing (AM). Traditional plastic hardening models (such as Hollomon, Ludwick, Voce, Swift, Ramberg-Osgood, and Swift-Voce) were used to describe the plastic region of stress–strain curves during tensile testing, but capturing their dependence on AM process settings is complex. The study trains neural networks to map key AM process parameters—including layer thickness, scanning speed, build orientation, laser power, hatch spacing, and laser spot size—as well as material properties (e.g., density, thermal conductivity) to strain-hardening responses. By comparing multiple hardening models through experimental validation, the work highlights the effectiveness of intelligent algorithms in modelling complex interactions and shows that the Ludwick model provided the best fit within the ANN framework. This approach offers a data-driven tool for predicting mechanical behaviour and aids optimization of AM processing for desired performance.