Collaborative Research: HCC: Medium: Co-Design of Shape and Fabrication Plans for Direct-Ink Write Printing Through Predictive Simulation

University Of Texas At Austin, Austin TX

Investigators

Abstract

Extrusion-based 3D printing is becoming an important tool in aerospace, automotive, medical, and defense applications, all of which require the highest confidence in their components. With the growth in the use of 3D-printed parts designed for cushioning, impact absorption, and integration of sensors and actuators, the importance of design tools that correctly account for and shape how a part is fabricated has never been more pronounced. Despite this need, current design tools focus exclusively on the part's outer geometric shape and ignore the printing process itself. In other words, the focus is on what is being printed rather than on how it's being printed, such as the path the printer nozzle takes when printing each layer of the part, despite the crucial effect the printing process has on the mechanical properties of the final part. This research will transform the design of 3D-printed parts by allowing users, for the first time, to design not only the part's shape but also its function. New algorithms developed by this project will automatically map from user-specified mechanical behavior to a Fabrication Plan that specifies the low-level details of how the printer should fabricate the part in order to achieve those goals. Since manufacturing applications excite students about STEM in an accessible, tangible way, this project will have additional broad impact by developing educational materials around the relationship between manufacturing and STEM careers, and by sharing project outcomes through established connections to local outreach efforts. This work will close the "form-function design gap" by creating new algorithms for co-designing both an object's macroscale shape and microscale fabrication plan, for 3D printers based on Direct-Ink-Write (DIW) technology. A microstructure-aware rod-based simulation will be developed to accurately analyze mechanical behavior of DIW-printed parts an order of magnitude more efficiently than traditional finite element methods. By allowing continuous variation of fabrication plans, the novel intermediate representation defined in this project will open the door for new interactive search and offline optimization strategies for additive manufacturing. Finally, the research will develop a novel bi-level optimization strategy that can jointly design object geometry and fabrication plan, while solving the challenge that each geometric design defines a different manifold of possible fabrication plans. The algorithms developed in this effort will be thoroughly tested against both mechanical experiments in the lab and simulated benchmarks. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.