Skip to Content
Machine Learning for Multiscale Thermal Simulation of Additive Manufacturing
/documents/17084224/17085097/PooriyaProject.png/7721ece8-99b6-45ce-aa27-824befd66f6e?t=1623446881050

Understanding the thermal histories evolving during the metal additive manufacturing process is the first step towards investigating the development of residual stresses, microstructure, and properties of manufactured parts. Thus, thermal simulation is crucial towards an improved understanding of the MAM process, and reliable models would be required to minimize the uncertainties that would be carried over to sequentially coupled mechanical and microstructural analyses.

The fast kinetics and highly localized nature of the phenomena involving melting and solidification of metal powder using a focused high-intensity heat source (e.g. laser) demand high levels of time and space discretization for MAM simulations which significantly increases the computational costs and makes part-scale calculations prohibitively expensive. Existing simplified simulation approaches apply gross approximations to overcome the numerical cost barrier and introduce high levels of uncertainty that ignores important process details. This project combines multiscale modelling and machine learning to enable part-scale high fidelity simulation and fast thermal analysis without compromising accuracy.  The efficiency and reliability of the proposed idea have been demonstrated for 2D thermal simulation [1,2]. This project is running in collaboration with Risk, Safety and Uncertainty Quantification and Seminar for Applied Mathematics from ETHZ.

Project Contact

 

 

Microstructure and Mechanical Property Prediction for Laser Powder Bed Fusion Process
/documents/17084224/17085097/JianProject.png/c1a5b07b-76a4-4b4f-8aee-ca591ba29d51?t=1623446864902

It is widely acknowledged that the microstructure of additively manufactured builds is considerably different to those observed in the cast or wrought alloys. AM microstructure is characterized by fine-elongated grains oriented in the build direction. Accordingly, anisotropic high strengths at room temperature and low creep performance at elevated temperatures were often observed for AM alloys. Parameters such as grain size, shape, crystallographic orientation and consequently, the mechanical properties were found to be very sensitive to the AM process parameters and therefore tuning the process conditions allows for the fabrication of builds with desired properties. Notably, the adoption of process conditions for different sites of a part enables the fabrication of functionally graded microstructures.

Full exploitation of the AM process potentials for designing the microstructure requires a deep understanding of microstructure development during the deposition process and its relationship with the resulted mechanical properties (i.e. process-microstructure-property correlation).  The running PhD project, in collaboration with the Computational Mechanics Group at ETH aims to develop a cellular automata-crystal plasticity finite element simulation framework for microstructure and mechanical properties simulation for AM process. Moderating the computational cost of such a detailed analysis through machine learning is planned.

Project Contact

Machine Learning for prediction of residual stresses and distortion in laser powder bed additive manufacturing
/documents/17084224/17085097/PatrikProject.png/fbf094e0-cc48-4f2f-84fa-1475cc050d89?t=1623446865108

While additive manufacturing enables the production of functionally graded materials and complex topologies, its employment to fabricate ready-to-use components is often inhibited by the induced distortions by the residual stresses. Steep temperature gradients and high cooling rates during the process and the resulted thermal strains induce residual stresses that can cause distortion and cracking of the built, affecting the geometrical accuracy and mechanical integrity of the products. To realise a 'first-time-right' high-quality production, a deeper understanding of the residual stress development process through high-fidelity numerical simulations and, therefore, adoption of residual stress and distortion mitigation strategies are required.  The high computational cost of such detailed simulations is a barrier against their adoption for the real-size components. Typically, gross simplifications at the expense of accuracy and reliability have been taken by the previous research for part-scale analyses.

After re-examing the common assumptions in the high-fidelity thermomechanical simulation (e.g. the employed constitutive model types), exploits the power of multiscale modelling and machine learning algorithms to realise reliable component-scale simulations for accurate prediction of distortion and residual stresses in AM components.

Project Contact

Neural Network Based Finite Element Method
/documents/17084224/17085097/IbrahimProject_302_203.png/e1cd6909-214d-4a41-b419-cf709c0bb59e?t=1622794628546

Additive manufacturing technologies offer enormous design freedom and enable the fabrication of highly complex geometries that would otherwise be too difficult or costly to be manufactured by traditional methods.  Numerical topology optimization is used to effectively exploit the offered design freedom for the fabrication of components that exactly meet the requirements of their application. Topology optimization of complex components based on the conventional finite element method is however computationally expensive and less effective for the design of multi-scale lattice structures which can be realised by additive manufacturing.

Within the ongoing PhD project, in collaboration with ZHAW and ETHZ, the development of a neural network-based finite element method is aimed for.  The methodology consists of an FE-model with super-element discretization capturing the global response of a multi-scale structure and a neural network-based surrogate model that provides local stiffness properties of the super-elements in dependence of the local material distribution (topology). This approach potentially allows to optimise multi-scale structures effectively and more efficiently. 

 

Project Contact

Place holder for projects from Abt. 204
/documents/17084224/17085097/302px.jpg/6df204b9-af2e-4a06-a324-ceba2bf54fa7?t=1622198467623

Additive manufacturing technologies offer enormous design freedom and enable the fabrication of highly complex geometries that would otherwise be too difficult or costly through traditional manufacturing methods.  Numerical topology optimization has a great contribution in effective exploitation of the offered design freedom for the fabrication of components that exactly meet the requirements of their application. Topology optimization of complex components based on conventional finite element method is however computationally expensive, due to the many required iterations.

Within the ongoing PhD project, in collaboration with ETHZ, the development of a neural network-based finite element method is aimed.  The methodology consists of an FE-model with super-element discretization capturing the global response of a multi-scale structure and a neural network-based surrogate model that provides local stiffness properties for the super-elements dependent on the local material distribution (topology) in the super-element.

Project Contact
Place holder for projects from Abt. 204
/documents/17084224/17085097/302px.jpg/6df204b9-af2e-4a06-a324-ceba2bf54fa7?t=1622198467623

Additive manufacturing technologies offer enormous design freedom and enable the fabrication of highly complex geometries that would otherwise be too difficult or costly through traditional manufacturing methods.  Numerical topology optimization has a great contribution in effective exploitation of the offered design freedom for the fabrication of components that exactly meet the requirements of their application. Topology optimization of complex components based on conventional finite element method is however computationally expensive, due to the many required iterations.

Within the ongoing PhD project, in collaboration with ETHZ, the development of a neural network-based finite element method is aimed.  The methodology consists of an FE-model with super-element discretization capturing the global response of a multi-scale structure and a neural network-based surrogate model that provides local stiffness properties for the super-elements dependent on the local material distribution (topology) in the super-element.

Project Contact

Quick Access

The latest EmpaQuarterly:
Focus: Energy revolution