• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>First European Workshop on Structural Health Monitoring 2002, Jul 10-12, 2002, Cachan (Paris), France >A Mathematical Network Approach to Structural Prognostic Health Management
【24h】

A Mathematical Network Approach to Structural Prognostic Health Management

机译：结构预测性健康管理的数学网络方法

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Evaluating the stresses of critical aircraft components is essential for advanced prognostic health monitoring. Aircraft stresses can be directly measured by strain gauges. Strain gauge monitoring systems only provide stress measurements at a small number of structural locations and their operational costs are often high. Recorded flight parameters can potentially indicate component stresses and at the same time describe flight conditions that caused these stresses, and hence, they can form the foundation of advanced prognostic management systems where the effects (stresses) and their causes can be derived from a single source. Over the last three decades, recorded flight data has been extensively used to analyse aircraft incidents and accidents and hence there is a growing interest in using readily available flight data for prognostic management. Therefore, Smiths Aerospace has developed mathematical networks that combine mathematical formulae, Artificial Intelligence (AI) and engineering knowledge to synthesis stresses from flight parameters. Working with BAE SYSTEMS, their engineering knowledge was used to configure mathematical networks for their aeroplanes. Whilst the networks accurately synthesised the stresses for a number of components, a challenge remained to define a method for qualifying them. This paper describes the AI model-based approach used, reports the results of the mathematical networks and presents preliminary discussions regarding the qualification approach of such networks.
机译：评估飞机关键部件的压力对于高级的预后健康监测至关重要。飞机应力可以通过应变仪直接测量。应变仪监控系统仅在少数结构位置提供应力测量，并且其运营成本通常很高。记录的飞行参数可以潜在地指示部件应力，同时描述导致这些应力的飞行条件，因此，它们可以构成高级预后管理系统的基础，在该系统中，影响（应力）及其原因可以从单一来源得出。在过去的三十年中，已记录的飞行数据已被广泛用于分析飞机的事故和事故征候，因此，人们对使用随时可用的飞行数据进行预后管理的兴趣日益浓厚。因此，史密斯航空航天局开发了数学网络，该网络将数学公式，人工智能（AI）和工程知识相结合，以综合来自飞行参数的应力。与BAE SYSTEMS合作，他们的工程知识被用来为其飞机配置数学网络。尽管网络可以准确地综合多个组件的应力，但是定义一种限定它们的方法仍面临挑战。本文描述了使用的基于AI模型的方法，报告了数学网络的结果，并对有关此类网络的资格方法进行了初步讨论。

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