首页> 外文期刊>IFAC PapersOnLine >Filtering and Smoothing of Hidden Monotonic Trends and Application to Fouling Detection ?
【24h】

Filtering and Smoothing of Hidden Monotonic Trends and Application to Fouling Detection ?

机译:过滤和平滑隐藏的单调趋势和应用于污垢检测

获取原文
           

摘要

In this paper, we present a filtering and smoothing scheme for process variables characterized by a hidden monotonic trend. The proposed method models the transition probability distribution of the hidden monotonic trend as a closed skew normal distribution and the observed data is assumed to have a Gaussian noise added onto this monotonic trend. The objective is to extract the monotonic trend given the noisy observations. The proposed method has advantages in process monitoring applications involving processes driven by a monotonic trend where vanilla Kalman filter may not be the apt option. The proposed method has been verified on an industrial dataset of a hot lime softener process to detect the fouling buildup.
机译:在本文中,我们提出了一种过滤和平滑方案,用于通过隐藏的单调趋势为特征的过程变量。 所提出的方法模拟隐藏单调趋势的转换概率分布作为闭合偏置的正态分布,并且假设观察到的数据具有添加到这种单调趋势上的高斯噪声。 目的是赋予嘈杂的观察来提取单调趋势。 该方法在过程监测应用中具有优势,涉及由Manilla Kalman滤波器可能不是APT选项的单调趋势驱动的过程。 所提出的方法已经在热石灰软化器工艺的工业数据集上验证,以检测污垢堆积。

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