Nonparametric Density Estimation of Hierarchical Probabilistic Graph Models for Assumption-Free Monitoring
Nonparametric Density Estimation of Hierarchical Probabilistic Graph Models for Assumption-Free Monitoring
复制标题
用于无假设监测的分层概率图模型的非参数密度估计
DOI：
10.1021/acs.iecr.6b04068
复制
发表时间：
2017-01
影响因子：
4.2
通讯作者：
Xie Lei
中科院分区：
文献类型：
--
作者：
Zeng Jiusun;Luo Shihua;Cai Jinhui;Kruger Uwe;Xie Lei
概率图模型，如贝叶斯网络，最近在过程监测和故障诊断中得到了关注。然而，它们的应用仅限于离散或连续高斯分布变量，这是由于难以有效地估计多元密度函数。这篇文章表明，将图形模型分解为层次结构可以将多变量密度函数的估计减少到低维/条件概率的估计。这些条件密度函数可以使用非参数核方法从数据中有效地估计，并且可以使用核密度估计（KDE）来估计低维密度。在估计的密度的基础上，异常过程行为可以通过检查哪个概率低于其相应的置信限来检测和诊断。模拟算例和工业高炉炼铁过程的应用表明，该方法是可行的。
Probabilistic graphical models, such as Bayesian networks, have recently gained attention in process monitoring and fault diagnosis. Their application, however, is limited to discrete or continuous Gaussian distributed variables, which results from the difficulty in efficiently estimating multivariate density functions. This article shows that decomposing the graphical model into a hierarchical structure reduces estimating a multivariate density function to the estimation of low-dimensional/conditional probabilities. These conditional density functions can be effectively estimated from data using a nonparametric kernel method and the low-dimensional densities can be estimated using a kernel density estimation (KDE). On the basis of the estimated densities, anomalous process behavior can be detected and diagnosed by examining which probability is lower than its corresponding confidence limit. Applications to simulated examples and an industrial blast furnace iron-making process show that the proposed metho...
登录
查看更多内容
DOI：
10.1016/s0169-7439(02)00140-5
发表时间：
2003-02
影响因子：
3.9
作者：
Leo H. Chiang;R. Braatz
通讯作者：
Leo H. Chiang;R. Braatz
影响因子：
3.7
作者：
Gao, Chuanhou;Chen, Jiming;Sun, Youxian
通讯作者：
Sun, Youxian
影响因子：
7.5
作者：
Friedman, N;Koller, D
通讯作者：
Koller, D
DOI：
10.3182/20070606-3-mx-2915.00004
发表时间：
2008-10
期刊：
IFAC Proceedings Volumes
影响因子：
--
作者：
Biao Huang
通讯作者：
Biao Huang
影响因子：
3.7
作者：
Jie Yu;Mudassir M. Rashid
通讯作者：
Jie Yu;Mudassir M. Rashid