• Matlab
• FEA
• additive manufacturing
• topology optimization
• structural optimization

Architectured materials are increasingly used in the design of light-weighted structures thanks to the progresses made in additive manufacturing. The Efficient Multi-scale Topology Optimization (EMTO) is developed for the 2D to build such structures by optimizing both the micro and macro scales. The EMTO uses a pre-computed database of micro-cells reducing the computational cost. To bridge the two scales, design variables, like the orientation of the cell, are defined. The results show a truss-like framework for the structures.
The objective of this thesis is to build a database of truss-like micro-cells, implementing a micro-scale structural optimization code. The new database is used in a slightly modified EMTO macro-scale optimization to verify if a reduction of the design space for the micro-cells is detrimental or not.
The cells are parametrized by superimposing well-defined beams whose thicknesses are controlled by the value assigned to the related parameters. The design conditions for the micro-cells are defined by the EMTO macro-scale variables. An inverse homogenization method based on the strain energy theorem is applied to get the properties of the cell and the optimality criteria method is used to solve the optimization problem. Once obtained the new database, several structural problems are solved.
Moreover, a generalization of the approach to the 3D is under developing. Limitations, issues and further implementations are investigated and described.