稳健性(进化)
离群值
算法
计算机科学
单变量
后验概率
一般化
贝叶斯概率
高斯过程
边际似然
数学
高斯分布
估计理论
贝叶斯推理
软传感器
拉普拉斯法
先验概率
推论
估计员
数学优化
人工智能
统计推断
模式识别(心理学)
多元正态分布
正态分布
概率逻辑
数据挖掘
贝叶斯网络
正态性
概率分布
作者
W. Hu,Yabin Zhang,Bowen Xu,Mingyu Dong,Tao Liu,Tianxu Hao,Min Liu
标识
DOI:10.1109/tim.2025.3636680
摘要
Various disturbances and uncertainties existing in actual industrial production environments can degrade the measurement accuracy of soft sensors based on single-value deterministic estimation. In addition, outliers caused by operational errors or recording mistakes may affect the generalization ability of soft sensors. Inspired by this, two Bayesian twin extreme learning machines based on symmetric skewed distributions, BTELM-ALD and BTELM-STD, are proposed. Both soft sensing methods perform parameter learning in a Bayesian framework, and train a pair of twin models based on combined weights to provide estimation intervals for key indicators. They employ skewed heavy-tailed distributions to model the residuals, which enhances robustness to outliers. BTELM-ALD utilizes an asymmetric Laplace distribution instead of Gaussian distribution and constructs a pair of twin models based on the combined weights (p,1-p). The introduction of suitable surrogate functions makes the posterior distribution and marginal likelihood easy to solve. In BTELM-STD, a univariate skewed t-distribution is presented and written as a hierarchical representation. The corresponding twin models are constructed based on the combined weights (s,-s), and then variational inference and the Newton method are utilized to optimize the parameters. Experimental results on several cases including an actual PTA oxidation process illustrate the validity and advantages of our proposed methods.
科研通智能强力驱动
Strongly Powered by AbleSci AI