CHS: Medium: Collaborative Research: Computational Design and 3D Printing of Textiles

纺织品不仅是世界时尚业的基本面料，也是汽车、航空、建筑和国防应用的重要部件。然而，纺织品原型和设计(无论是服装、室内装潢还是复合材料)是一个艰巨而昂贵的过程。这项研究项目旨在了解和推进新的添加剂制造技术(通常称为“3D打印”)在纺织产品设计和原型制作中的作用。PI的目标是开发3D打印硬件和计算机软件，使工程师能够更快、更经济地制作纺织品设计原型，并更好地控制广泛的机械、光学和电气特性，如空气动力学阻力、附着力、热调节、摩擦、弹性、孔隙率、密度、导电性和视觉外观。除个别纺织品外，项目成果将支持制造不需要大量缝合和组装的完整产品，其中可能包括难以从平板切割的曲面形状和/或成本太高而无法通过传统方法制造的复杂复合材料组件。为了实现这些目标，采购经理人将开发：一个高度优化的纺织品“单元”库，它可以使用一种新的纺织功能语言组合起来，形成大量可能的纺织品；计算机优化软件，可以精确控制纺织品特性；一个计算机程序，允许用户直观和交互地设计复杂的纺织品；以及一个专门的3D打印机，能够精确地制造涉及多种材料的纺织品。从技术上讲，这个项目将创建第一个完整的硬件/软件管道，用于使用多材料3D打印进行纺织品的数字化设计和制造。这条管道中的第一个基本步骤是构建参数化元材料模板，该模板为用户提供用于调整纺织品行为和大规模属性的高级旋钮。下一步，将通过将连续统同质化和数据驱动方法相结合来实现交互模拟虚拟纺织品的行为的能力；PI将开发一种交互设计工具，该工具采用与物理模拟绑定的一阶灵敏度分析，使设计师能够在微观和宏观层面上导航可能设计的巨大空间。采用减速机-调谐器模型的一种新的功能指定纺织品设计语言将允许工程师和设计师根据所需的行为和性能指定元材料，从而使设计能够保证其特性和符合标准。目前3D打印机的打印量是有限的；然而，通过将计算纺织品折叠纳入管道中，PI的系统将能够以更小的折叠配置打印非常大的设计。折叠问题的求解涉及到具有单向接触约束的非线性、非凸、最优化问题。最后，纺织品和服装将使用现成的3D打印机和一个新型的低成本、高分辨率、模块化的3D打印平台打印，该平台能够打印多达12种不同的材料，这些材料在机械和外观特性上存在差异。除了光致聚合物材料，PI还计划将硬件能力扩展到使用共聚物和溶剂型材料的3D打印结构。有关该项目的更多信息，请访问http://textiles.csail.mit.edu/

In addition to being the essential fabric of the world's fashion industry, textiles are important components for automotive, aeronautical, architectural, and defense applications. Yet textile prototyping and design (whether for garments, upholstery or composite materials) is an arduous and expensive process. This research project seeks to understand and advance the role of new additive manufacturing technologies (commonly referred to as "3D printing") in the design and prototyping of textile products. The PIs' goal is to develop 3D printing hardware and computer software that enable engineers to prototype textile designs more quickly and economically, and with greater control over a broad gamut of mechanical, optical, and electrical characteristics such as aerodynamic drag, adhesion, heat regulation, friction, elasticity, porosity, density, electrical conductivity, and visual appearance. Beyond individual textiles, project outcomes will support the fabrication of complete products that do not require considerable stitching and assembly, and which may include curved shapes too difficult to cut from flat panels and/or complex composite assemblies too costly to fabricate via traditional methods. To achieve these objectives, the PIs will develop: a library of highly-optimized textile "units" that can be combined using a new language of textile functionality to form a vast array of possible textiles; computer optimization software that enables precise control of textile properties; a computer program that allows users to visually and interactively design complex textile products; and a specialized 3D Printer that is able to precisely fabricate textiles involving multiple materials.Technically speaking, this project will create the first complete hardware/software pipeline for digital design and fabrication of textiles using multi-material 3D printing. The first fundamental step in this pipeline is constructing parameterized meta-material templates that provide users with high-level knobs for tuning the behavior and large-scale properties of a textile. Next, the ability to interactively simulate the behavior of a virtual textile will be achieved by combining continuum homogenization and data-driven methods; the PIs will develop an interactive design tool that employs first order sensitivity analysis tied to the physical simulation, to enable designers to navigate the huge space of possible designs at both the micro and macro levels. A new language for functionally specifying textile designs that employs a reducer-tuner model will allow engineers and designers to specify meta-materials in terms of desired behavior and performance, enabling designs with guarantees on their characteristics and compliance with standards. Printing volumes for current 3D printers are limited; however, by incorporating computational textile folding into the pipeline, the PIs' system will be able to print very large designs in much smaller folded configurations. Solution of the folding problem will involve nonlinear, non-convex, optimization with unilateral contact constraints. Finally, textiles and garments will be printed using both off-the-shelf 3D printers and a novel low-cost, high-resolution, modular 3D printing platform that is capable of printing with up to 12 different materials that vary in mechanical and appearance properties. In addition to photopolymer materials, the PIs plan to extend hardware capabilities to 3D print structures using co-polymers and solvent-based materials. More information about this project is available online at http://textiles.csail.mit.edu/