Research Projects

Laser additive manufacturing has shown considerable promise for diverse applications in the field of aerospace, automobile engineering, biomedicine, etc. due to the ability of near net shape fabrication of complex structures with customized geometries. However, to obtain high quality products, it is essential to optimize the microstructure formed during solidification of the metal or alloy during the laser melting process. In view of this, the proposed project aims to develop a comprehensive simulation tool which combines multiple length scales (system level and microstructure level). The proposed project involves the development of a comprehensive computational tool for multi-scale simulation of the two main types of laser based additive manufacturing processes: (a) Direct Metal Deposition (DMD) & (b) Selective Laser Melting (SLM) / Laser Powder Bed Fusion. The tool will be developed in modular form with common modules for heat transfer, fluid flow, melting & solidification, grain nucleation & grain growth, species transport, etc. for phenomena common for laser additive manufacturing in general. Separate modules will be implemented for process specific details for DMD & SLM such molten droplet addition during layer formation in DMD and melting in a porous bed in SLM. The model will also include for formation of multiple layer and multiple track formation. - The project is planned to be done in two phases (3 years + 2 years) and will involve the following work. [A] Phase – I (years 1-3) 1. Development of simulation platform – Implementation of physics based modules – heat transfer & phase change with accurate interface tracking, dendrite nucleation & growth, overall microstructure evolution, laser-powder interaction, Marangoni & buoyancy flow, formation of multiple layers and multiple tracks, species transport. Separate modules for process specific details for DMD & SLM such as molten droplet addition during layer formation in DMD and melting in a porous bed in SLM. All the models will be implemented using Fortran with a base finite volume solver integrated with all the modules. Thus there will be no dependence on commercial software. 2. Experimental analysis for in-house DMD and commercial SLM for 3 sets of highly relevant alloys (Al-Cu, Ti, Inconel) 3. Validation of the developed simulation platform using the generated experimental results. [B] Phase – II (years 4-5) • In the 2nd phase of the project, a Deep Learning (DL) based surrogate model will be developed for prediction of the effects of process parameters on the final microstructure. Subsequently, an optimization module will be developed and integrated with the computational platform. The optimization module will use the developed DL based model to predict the most suitable process parameters for desired microstructure and mechanical properties. • In addition to the developed numerical solver, a graphical user interface (GUI) will be developed which will help the user to specify the simulation parameters and enable the necessary modules easily without modifying the source code • It is planned that in future, the computational model will be connected with a materials property database and thermodynamic phase diagram calculation tool. However, this is kept outside the scope of the proposed 5 year project. The developed tool will directly advance the state of the art in modelling and simulation of laser additive manufacturing, in particular (a) direct metal deposition (b) laser powder bed fusion processes, by providing process level modelling capability combined with microstructure & individual grain scale resolution.