Strength based optimization and test of fiber reinforced additively manufactured structures considering topology and fiber orientation
Authors
Gadegaard, Frederik Juel ; Thuesen, Jan
Term
4. term
Education
Publication year
2022
Abstract
This thesis explores how fiber-reinforced FFF (Fused Filament Fabrication) 3D printing can be combined with simultaneous topology and fiber orientation optimization to design strong, material-efficient structures. The objective is to minimize volume subject to a strength constraint using the Tsai-Wu failure criterion, with a linear elastic material model. Material properties are experimentally determined for two printed materials—a short fiber-reinforced polyamide and a polyamide with continuous fibers—and used in the optimization. The method couples density-based topology optimization with Discrete Material Optimization (DMO) for fiber angles. Because the original DMO formulation over-penalizes intermediate densities and is incompatible with density filtering, two remedies are proposed and implemented: (1) reducing penalization by avoiding local summation of failure indices and (2) removing filtering, which necessitates a quadratic element formulation to prevent checkerboarding. Both approaches are studied numerically and validated experimentally for the short fiber-reinforced material; the results show a conservative (underestimated) structural strength, attributed to post-processing that adds material for printability. Both methods are viable, but the reduced-penalization approach with filtering is recommended, as it introduces a minimum length scale and yields more robust, fabrication-ready designs.
Specialet undersøger, hvordan fiberforstærket FFF (Fused Filament Fabrication) 3D-print kan kombineres med samtidig topologi- og fiberretningsoptimering for at designe stærke, materialeeffektive strukturer. Målet er at minimere volumen under en styrkebetingelse baseret på Tsai-Wu svigtkriteriet, hvor materialemodellen antages lineært elastisk. Materialeparametre bestemmes eksperimentelt for to 3D-printmaterialer: en kortfiberforstærket polyamid og en polyamid forstærket med kontinuerte fibre, og disse data anvendes i optimeringen. Metodisk kombineres densitetsbaseret topologioptimering med Discrete Material Optimization (DMO) til fiberretninger. Da den oprindelige DMO-formulering straffer mellemtætheder for hårdt og ikke kan bruges sammen med densitetsfiltrering, foreslås og implementeres to løsninger: (1) reducere penaliseringsgraden ved ikke at summere svigtindeks lokalt og (2) fjerne filtrering, hvilket kræver en kvadratisk elementformulering for at undgå numeriske artefakter (checkerboarding). Begge metoder afprøves i numeriske eksempler og valideres eksperimentelt for det kortfiberforstærkede materiale; resultaterne indikerer en konservativ (underestimeret) styrke, hvilket tilskrives nødvendig efterbehandling, der tilføjer materiale til 3D-print. Begge tilgange vurderes anvendelige, men metoden med reduceret penaliseringsgrad og bevaret filtrering anbefales, da den indfører en minimumslængdeskala og giver mere robuste, fremstillingsparate løsninger.
[This apstract has been generated with the help of AI directly from the project full text]