Targeted Structural Optimization with Additive Manufacturing of Metals

The recent advances in Additive Manufacturing (AM) of metals have now improved the state-of-the-art such that traditionally non-producible parts can be readily produced at in a cost effective way. Because of these advances in manufacturing technology, structural optimization techniques are well positioned to supplement and advance this new technology. The goal of this proposal is to develop a structural design, analysis and optimization framework combined with AM to significantly lightweight the interior of metallic structures, while maintaining the selected structural properties of the original solid. This is a new state-of-the-art capability to significantly reduce mass, while, maintaining the structural integrity of the original design is something that can only be done with AM. In addition, this framework will couple the design, analysis and fabrication process, meaning that what has been designed directly represents the produced part, thus closing the loop on the design cycle and removing human iteration between design and fabrication. This fundamental concept has applications from light weighting launch vehicle components to in-situ resource fabrication. The outcome of this proposal will results in the following deliverables: Assess and implement the proposed optimization concept Print and test one optimized sample part and collect strength data Provide comparisons to other stiffened structures concepts Explore applicable material failure models (research that is being guided by the Ames funded effort) The current state of the art is that we can produce sample parts with Additive Manufacturing of metals. However, the material properties are still such an unknown, that this process can't be fully utilized for load bearing components. These material properties are going to be researched by a recently funded Ames effort. Once this is mature, NASA will then have the capability to engineer AM parts. Once this is possible we can begin taking advantage of a whole new design philosophy that is now possible with AM and utilize structural optimization to its fullest. This will allow lightweight, metallic parts to be manufactured, cheaply. If the research presented in this proposal is funded now, this new, innovative design process will be mature and ready to be implemented with material properties research that is in work at Ames. This design automation and optimization framework is akin to a field of research known as topology optimization combined with a variation of shape optimization. Topology optimization is often used to minimize the mass required by a part, while maintaining the overall stiffness required for the structure by removing unloaded material. This area has seen a plethora of attention in the literature and academia and has been implemented in many commercial applications in the general sense. The research activity proposed here effectively combines topology with Additive Manufacturing resulting in minimal mass hardware and furthers this area by using Adaptive Mesh Refinement with shape optimization which is a relatively new area of research. The process is simple in thought; typically structural analysis utilizes the Finite Element Method. This process discretizes a continuum structures into discrete solid elements to perform an analysis that will determine the overall strength of the structure. The core of the idea is this. What if using these solid elements, the solid fill could be removed, but the defining faces of that solid remain as shells? The part will now be lighter, with a similar overall strength and stiffness. This is essentially the core concept behind grid stiffened structures, this time the outer-profile remains intact to maintain form, fit and function. In traditional manufacturing, this concept is impossible to implement without joining panels adding complexity and cost, but AM can easily achieve this. The process begins with a part that has already been designed to meet form, fit and function and is considered to be mature in the design cycle, meaning all applicable requirements (loads, environments, etc.) have been estimated. The first step in the optimization is to produce a 3-dimensional Finite Element Model (FEM) using tetrahedral elements and run an analysis to determine where material is not being fully utilized. From there, the optimization will begin removing material by converting tetrahedral elements to 2 dimensional shell elements with a minimal thickness, thus hollowing the element and removing inefficiently utilized mass. The newly formed shell model is then re-run in the analysis and assessed and ranked based upon and objective function. This loop is repeated until convergence is met. A 3 dimensional Computer Aided Design model of the optimized internal topology is then generated automatically. The outer mold and new internal topology are passed to the Additive Manufacturing process to be produced. More »