NIRT: Modeling and Simulation Framework at the Nanoscale. Application to Process Simulation, Nanodevices, and Nanostructured Composites

NIRT：在Nanoscale的建模和仿真框架摘要本提案是响应纳米科学与工程倡议，NSF 02-148，类别NIRT。 研究工作的目标是发展一个通用的多尺度建模和模拟方法的纳米结构材料系统。项目工作将包括：（一）开发一种数学上严格的多尺度建模方法，能够将原子尺度的行为与全尺度系统相结合，（二）围绕这种方法建立一个计算模拟框架，可以有效地将各种尺度的行为调查技术整合到该框架中，以及（iii）使用与伦斯勒目前正在进行的实验工作的协同相互作用，对已开发的核心技术进行概念验证。该项目将汇集具有互补专业知识的研究人员：纳米力学，多体动力学，多尺度计算技术和计算工程。基于自适应分层建模的形式主义的纳米级建模和仿真的可靠性是该项目的关键方面。努力将集中在耦合的建模方法应用在原子和连续的水平，包括适当的时间尺度和整合各种物理现象的桥梁。 将特别考虑建模误差的估计。这些估计将基于后验测量，这些后验测量将用于新的程序，以适应性地选择在整个模拟空间/时间域应用的尺度和方法。开发的核心技术将被纳入一个基于组件的仿真框架。建立在先进的软件技术，该框架将有效地支持分布式并行计算的自适应多尺度模拟的不断发展的结构。为了解决这些模拟的计算要求，将开发新的数值方法，减少这些模拟的计算工作量。这一努力预计将影响科学和工业的建模，分析和理解应用中使用的大量纳米材料的能力，例如发动机，航空航天和汽车工业中使用的轻质部件以及耦合的机电设备（传感器）。这些应用代表了美国主要行业的技术。需要坚实的理论基础和相关的模拟能力来支持这些行业增长的突破性发展。该项目将利用伦斯勒战略计划支持的教师，课程和设施。该计划还将开发新的跨学科研究生和短期课程，在多尺度建模与应用纳米技术，以培养一代科学家和工程师与上级数学和技术技能。

NIRT: Modeling and Simulation Framework at the NanoscaleAbstract This proposal was received in response to Nanoscale Science and Engineering initiative, NSF 02-148, category NIRT. The goal of the research effort is the development of a generalized multiscale modeling and simulation methodology for nanostructured material systems. Project efforts will include the development of (i) a mathematically rigorous multiscale modeling methodology capable of coupling behaviors from the atomic scale through full scale system, (ii) a computational simulation framework built around this methodology into which techniques for investigating behaviors at the various scales can be effectively integrated, and (iii) a proof of concept of the developed core technologies using synergetic interactions with experimental work currently under way at Rensselaer. The project will bring together researchers with complementary expertise in: nanomechanics, multibody dynamics, multiscale computational techniques, and computational engineering.Reliability of modeling and simulation at the nanoscale based on the formalism of adaptive hierarchical modeling is the key aspect of the project. Efforts will focus on the coupling of modeling methods applied at the atomic and continuum levels, including bridging of the appropriate time scales and integrating various physical phenomena. Specific consideration will be given to the estimates of modeling errors. These estimates will be based on a posteriori measurements that will be employed in new procedures for the adaptive selection of scales and methods applied throughout the space/time domain of the simulation. The core technologies developed will be incorporated into a component-based simulation framework. Building on advanced software technologies, this framework will efficiently support distributed parallel calculations on the evolving structures of adaptive multiscale simulations. To address the computational requirements of these simulations new numerical methods that reduce the computational effort of these simulations will be developed. This effort is expected to impact science and industry's ability to model, analyze, and understand a vast array of nanomaterials used in applications, such as engines, lightweight components used in the aerospace and automotive industries, and coupled electro-mechanical devices (sensors). These applications represent enabling technologies for major U.S industries. Solid theoretical foundations and associated simulation capabilities are needed to support the breakthrough developments central to the growth of these industries. This project will take advantage of the faculty, programs and facilities being supported through the Rensselaer Strategic Plan. The program will also develop new interdisciplinary graduate and short courses in multiscale modeling with applications to nanotechnology in order to train a generation of scientists and engineers with superior mathematical and technological skills.