Novel Just-In-Time Learning-Based Soft Sensor Utilizing Non-Gaussian Information

Novel Just-In-Time Learning-Based Soft Sensor Utilizing Non-Gaussian Information
复制标题

利用非高斯信息的新型即时学习软传感器

DOI:
10.1109/tcst.2013.2248155
复制
发表时间:
2014
影响因子:
4.8
通讯作者:
Gao, Chuanhou
Gao, Chuanhou
中科院分区:
计算机科学2区
文献类型:
--
作者:
Xie, Lei;Zeng, Jiusun;Gao, Chuanhou

文献摘要

参考文献

被引文献

相似文献

本简介开发了一种新颖的基于即时 (JIT) 学习的软传感器,用于工业流程建模。假设记录的数据表现出非高斯信号分量,这些分量是通过非高斯回归(NGR)技术提取的。与之前使用基于距离的相似性度量进行局部建模的 JIT 建模工作不同,本文介绍了一种使用支持​​向量数据描述提取的非高斯分量的新相似性度量。基于相似性度量,提出了称为 NGR_JIT 的 JIT 建模过程。对数值示例以及工业过程的应用研究表明,当预测变量和响应集是非高斯分布时,所提出的软传感器可以提供更好的预测精度。
This brief develops a novel just-in-time (JIT) learning-based soft sensor for modeling of industrial processes. The recorded data is assumed to exhibit non-Gaussian signal components, which are extracted by a non-Gaussian regression (NGR) technique. Unlike previous work on JIT modeling which uses distance-based similarity measure for local modeling, this brief introduces a new similarity measure for the extracted non-Gaussian components using support vector data description. Based on the similarity measure, a JIT modeling procedure called NGR_JIT is proposed. Application studies on a numerical example as well as an industrial process demonstrate the proposed soft sensor can give better predictive accuracy when the predictor and response sets are non-Gaussian distributed.
DOI: 10.1016/s0169-7439(98)00024-0
发表时间: 1998-07-06
影响因子: 3.9
作者:
Rannar, S;MacGregor, JF;Wold, S
通讯作者: Wold, S
基于即时学习的在线软测量建模方法的比较研究
DOI: 10.1016/j.chemolab.2010.09.008
发表时间: 2010-12-15
影响因子: 3.9
作者:
Ge, Zhiqiang;Song, Zhihuan
通讯作者: Song, Zhihuan
DOI: 10.1023/b:mach.0000008084.60811.49
发表时间: 2004-01-01
期刊: MACHINE LEARNING
影响因子: 7.5
作者:
Tax, DMJ;Duin, RPW
通讯作者: Duin, RPW
DOI: 10.1016/s0959-1524(99)00052-9
发表时间: 2000-12
影响因子: 4.2
作者:
V.R Radhakrishnan;A.R Mohamed
通讯作者: V.R Radhakrishnan;A.R Mohamed
DOI: 10.1142/s0129065700000028
发表时间: 2000-02-01
影响因子: 8
作者:
Bingham, E;Hyvarinen, A
通讯作者: Hyvarinen, A