Loading…

Explainable fault diagnosis of gas-liquid separator based on fully convolutional neural network

•An explainable diagnosis network based on FCN is designed for gas-liquid separator.•The class activation mapping is considered to ensure the explainability of feature extraction.•The output of the global average pooling layer is applied to analyze the engineering significance of convolutional kerne...

Full description

Saved in:
Bibliographic Details
Published in:Computers & chemical engineering 2021-12, Vol.155, p.107535, Article 107535
Main Authors: Liu, Jiaquan, Hou, Lei, Wang, Xin, Zhang, Rui, Sun, Xingshen, Xu, Lei, Yu, Qiaoyan
Format: Article
Language:English
Subjects:
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:•An explainable diagnosis network based on FCN is designed for gas-liquid separator.•The class activation mapping is considered to ensure the explainability of feature extraction.•The output of the global average pooling layer is applied to analyze the engineering significance of convolutional kernels.•The developed method can be well applied to other production equipment. The diagnosis of slug flow is extremely important for the efficient operation of the gas-liquid separator. Traditional fault diagnosis based on the convolutional neural network has not involved the explainability of the convolutional neural network, which makes the model difficult to interpret from the perspective of physical meaning. An explainable diagnostic method based on a fully convolutional neural network is proposed. The class activation mapping, multivariate mutual information, global average pooling and t-distributed stochastic neighbor embedding are combined to analyze the diagnostic process of the network. The experimental results based on simulation data showed that the proposed method can be utilized to interpret the correlation degree between different operating conditions, the importance of each period in the measurement variable, and the engineering significance of the convolutional kernels of the last layer, which provides information supplement for fault reasoning of human experts.
ISSN:0098-1354
1873-4375
DOI:10.1016/j.compchemeng.2021.107535