文献类型：期刊文献

英文题名：Adaptive Manifold Discriminative Distribution Alignment for Fault Diagnosis of Chemical Processes

作者：Xiao, Yutang[1];Shi, Hongbo[1];Wang, Boyu[2,3];Tao, Yang[1];Tan, Shuai[1];Song, Bing[1]

机构：[1]East China Univ Sci & Technol, Minist Educ, Key Lab Smart Mfg Energy Chem Proc, Shanghai 200237, Peoples R China;[2]Univ Western Ontario, Dept Comp Sci, London, ON N6A 5B7, Canada;[3]Univ Western Ontario, Brain Mind Inst, London, ON N6A 5B7, Canada

年份：2021

卷号：60

期号：27

起止页码：9860

外文期刊名：INDUSTRIAL & ENGINEERING CHEMISTRY RESEARCH

收录：;EI(收录号：20213010672114);WOS:【SCI-EXPANDED(收录号:WOS:000674325100020)】；

基金：This research is supported by the National Natural Science Foundation of China (Nos. 62073140 and 62073141) and the National Natural Science Foundation of Shanghai (No. 19ZR1473200). B.W. is supported by the NSERC Discovery Grants Program.

语种：英文

外文关键词：Fault detection - Numerical methods - Probability distributions

摘要：Fault diagnosis plays a significant role in chemical processes to avoid serious accidents that challenge production safety. In practice, however, to establish fault diagnosis models, it is expensive to collect labeled fault data, since a fault is a small probability event in a chemical process. Consequently, it is challenging to build a reliable fault diagnosis model, if a process has a small amount of fault data. On the other hand, we may have sufficient data collected from other processes, which can be leveraged using domain adaptation techniques. However, traditional domain adaptation approaches use only samples to align the statistical distribution without using the label information, and focus only on aligning the input features of the data to improve their transferability yet ignoring their discriminability. As a result, the model trained using such an approach may still have poor performance. To address this issue, we propose a novel adaptive manifold discriminative distribution alignment (AMDDA) approach to align the statistical distribution between domains in manifold space based on the discriminative conditional probability, in which the transferability and discriminability are captured simultaneously. In addition, by leveraging the clustering assumption of domain adaptation, AMDDA adopts an adaptive pseudo-label updating strategy to improve the quality of pseudo-label during the training process. Empirical results of both the numerical example and the Tennessee Eastman process demonstrate the effectiveness and efficiency of our method.