• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>IEEE transactions on industrial informatics >Probability Analysis for Failure Assessment of Electric Energy Metering Equipment Under Multiple Extreme Stresses
【24h】

Probability Analysis for Failure Assessment of Electric Energy Metering Equipment Under Multiple Extreme Stresses

机译：多重应力下电能计量设备失效评估的概率分析

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

The failure evaluation of electric energy metering equipment is essential for the equipment design and accurate measurement of electric energy, especially in extreme environmental stress. However, actual failure assessment is often affected by the environmental noise and insufficient interpretability. To address this problem, this article first proposes an improved k-nearest neighbor (IkNN) to identify potential outliers. In addition, an optimized distance function is used to obtain the score for each outlier. Next, a probability analysis method, namely, the weighted fusion Bayesian (WFB), is proposed to fuse multiple extreme environmental stresses and failure rate using the proposed nonlinear fusion function. Combining the WFB and the IkNN, examples from three extreme environmental regions show that the proposed evaluation framework has a higher assessment performance and less uncertainty. Compared with the classical prediction methods, our framework has profound outlier detection and failure prediction performance ever under the condition of small samples. More importantly, the parameters of this model are interpretable compared to some conventional approaches.
机译：电能计量设备的故障评估对于设备设计和精确测量电能来说是必不可少的，尤其是极端环境压力。但是，实际的失败评估往往受环境噪音的影响，并且无法解释性不足。为了解决这个问题，本文首先提出了一个改进的K-最近邻居（IKNN）来识别潜在的异常值。此外，优化的距离功能用于获得每个异常值的分数。接下来，提出了一种概率分析方法，即加权融合贝叶斯（WFB），使用所提出的非线性融合功能熔断多个极端环境应力和失效率。组合WFB和IKNN，三个极端环境区域的例子表明，所提出的评估框架具有更高的评估性能和不确定性。与经典预测方法相比，我们的框架在小样本的条件下具有深刻的异常检测和故障预测性能。更重要的是，与一些传统方法相比，该模型的参数是可解释的。

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