SBIR Phase I: Modeling of Thermal Transport and its Interaction with Crystal Formation in Optical Fiber Drawing on Distributed Memory Machines

这个小企业创新研究第一阶段项目是开发和演示一个计算工具，用于在分布式存储机器上详细模拟光纤拉伸过程中的热传输。ZBLAN光纤具有更宽的光谱和更低的损耗系数，有可能在下个世纪取代现有的二氧化硅光纤。然而，ZBLAN玻璃在光纤拉伸过程中容易结晶。抑制结晶需要清楚地了解热传递，包括辐射传热和弯曲自由表面的两相流。在第一阶段，将修改有限体积法（FVM）来模拟气体外壳和玻璃内部的辐射传热。气-玻璃界面被视为一个光学定向和反射表面。在界面处的共轭边界条件耦合下，求解了玻璃和外部气体的全椭圆控制方程。物理域的离散化将通过多区域自适应网格生成（MAGG）技术进行。在求解过程中，通过消息传递接口（MPI）库开发并实现了一种高效的并行算法。本文将开发一种并行代数多网格求解器来求解这些离散方程。第一阶段将展示所提出的热传输过程模拟工具的高精度和高效率。在第二阶段，将建立晶体形成模型，并在工具中完成热输运与晶体形成之间的耦合。要开发的仿真工具将大大有利于光纤工业，需要详细了解多模和高耦合输运现象及其与热致缺陷的相互作用。潜在的应用包括设计、优化和控制光纤拉伸过程和许多其他制造和材料加工系统。关键词：光纤拉深，辐射传热，并行计算。

This Small Business Innovation Research Phase I project is to develop and demonstrate a computational tool for detailed simulation of thermal transport in a optical fiber drawing process on distributed memory machines. The ZBLAN optical fibers may replace the existing silica optical fibers in the next century because they have broader spectrum and lower loss coefficient. However, the ZBLAN glass tends to crystallize during the drawing of optical fiber. The suppression of crystallization requires a clear understanding of thermal transport involving radiative heat transfer and two-phase flow with a curved free surface. In Phase I, the finite volume method (FVM) will be modified to simulate radiative heat transfer in the gas enclosure as well as inside the glass. The gas-glass interface is treated as an optically directional and reflecting surface. The full elliptic governing equations will be solved for both glass and external gas, which are coupled by the conjugate boundary conditions at the interface. The discretization of the physical domain will be carried out by the multizone adaptive grid generation (MAGG) technique. An efficient parallel algorithm will be developed and implemented in the solution procedure with message passing by the Message Passing Interface (MPI) library. A parallel algebraic multi-grid (AMG) solver will be developed to solve the discretized equations. The Phase I will demonstrate the high accuracy and efficiency of the proposed simulation tool for thermal transport process. In Phase II, the crystal formation model will be developed, and the coupling between thermal transport and crystal formation will be completed in the tool. The simulation tool to be developed will significantly benefit the optical fiber industry that requires a detailed understanding of multimode and highly coupled transport phenomena and their interactions with thermal induced defects. The potential applications include the design, optimization, and control of optical fiber drawing process and many other manufacturing and materials processing systems.Key Words: Optical Fiber Drawing, Radiative Heat Transfer, Parallel Computing.