• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Reliability Engineering & System Safety >A novel learning function based on Kriging for reliability analysis
【24h】

A novel learning function based on Kriging for reliability analysis

机译：基于Kriging的新型学习功能用于可靠性分析

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

Adaptively constructing the surrogate model for reliability analysis has been widely studied for the advantage of guaranteeing the estimation accuracy while calling the real performance function as little as possible. A new learning function called Folded Normal based Expected Improvement Function (FNEIF) is proposed to efficiently estimate the failure probability. Firstly, an improvement function is constructed by treating the prediction of surrogate model as folded normal variable, while the expectation function of the folded normal variable is an excellent index for measuring the contribution of a point to improve the surrogate model. Secondly, the expectation of the improvement function is analytically derived to identify the new training sample. Thirdly, a new stopping criterion is established based on the uncertainty magnitude of the prediction. Numerical and engineering application examples are introduced to show the effectiveness of the proposed learning function FNEIF for reliability analysis.
机译：为了保证估计精度，同时尽可能少地调用真实性能函数的优点，已经广泛研究了自适应地构建用于可靠性分析的替代模型。提出了一种新的学习功能，称为基于折叠法线的预期改进功能（FNEIF），可以有效地估计故障概率。首先，通过将替代模型的预测作为折叠法线变量来构造改进函数，而折叠法线变量的期望函数是衡量改进替代点的贡献的极好的指标。其次，通过分析得出改进功能的期望值以识别新的训练样本。第三，基于预测的不确定性幅度，建立了新的停止准则。通过数值和工程应用实例介绍了所提出的学习函数FNEIF对可靠性分析的有效性。

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