• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International Topical Meeting on Nuclear Reactor Thermal Hydraulics >GEOMETRIC UNCERTAINTY QUANTIFICATION AND ROBUST DESIGN FOR 2D SATELLITE SHIELDING
【24h】

GEOMETRIC UNCERTAINTY QUANTIFICATION AND ROBUST DESIGN FOR 2D SATELLITE SHIELDING

机译：2D卫星屏蔽的几何不确定性量化和鲁棒设计

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

The design of satellites usually includes the objective of minimizing mass due to high launchcosts, which is challenging due to the need to protect sensitive electronics from the spaceradiation environment by means of radiation shielding. This is further complicated by the needto account for uncertainties, e.g. in manufacturing. There is growing interest in automateddesign optimization and uncertainty quantification (UQ) techniques to help achieve thatobjective. Traditional optimization and UQ approaches that rely exclusively on responsefunctions (e.g. dose calculations) can be quite expensive when applied to transport problems.Previously we showed how adjoint-based transport sensitivities used in conjunction withgradient-based optimization algorithms can be quite effective in designing mass-efficientelectron and/or proton shields in one- or two-dimensional Cartesian geometries. In this paperwe extend that work to UQ and to robust design (i.e. optimization that considers uncertainties)in 2D. This consists primarily of using the sensitivities to geometric changes, originallyderived for optimization, within relevant algorithms for UQ and robust design. We performUQ analyses on previous optimized designs given some assumed manufacturing uncertainties.We also conduct a new optimization exercise that accounts for the same uncertainties. Ourresults show much improved computational efficiencies over previous approaches.
机译：卫星的设计通常包括由于高发射而最小化质量的目标由于需要保护敏感电子器件从空间中保护敏感电子产品而具有挑战性辐射环境通过辐射屏蔽。这对需要进一步复杂化要考虑不确定性，例如，在制造中。对自动化的兴趣日益增长设计优化和不确定性量化（UQ）技术帮助实现这一目标客观的。传统优化和UQ接近依赖于响应的方法功能（例如剂量计算）在应用于运输问题时可以非常昂贵。以前我们展示了与之结合使用的基于伴随的运输敏感性基于梯度的优化算法在设计批量效率方面非常有效电子和/或质子屏蔽在一个或二维笛卡尔几何形状中。在本文中我们将其延伸到UQ并强制设计（即，考虑不确定性的优化）在2D。这主要包括使用敏感性对几何变化，最初在UQ和鲁棒设计的相关算法中导出优化。我们表演UQ分析先前的优化设计，给出了一些假设的制造不确定性。我们还开展了一个新的优化练习，占同样的不确定性。我们的结果显示出对先前方法的改进改善的计算效率。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号