TY - JOUR
T1 - A Novel Deep Learning-Based Robust Dual-Rate Dynamic Data Modeling for Quality Prediction
AU - Meng, Xiangan
AU - Liu, Qiang
AU - Yang, Chao
AU - Zhou, Le
AU - Cheung, Yiu-Ming
N1 - Publisher Copyright:
© 2023 IEEE
Funding Information:
This work was supported in part by the National Natural Science Foundation of China (61991401, 62161160338, U20A20189), National Natural Science Foundation of
China (NSFC)/Research Grants Council (RGC) Joint Research Scheme N HKBU214/21, the General Research Fund (GRF) of RGC with the grant number: 12202622, and 111 Project 2.0 (No. B08015).
PY - 2023/5/12
Y1 - 2023/5/12
N2 - Traditional data-driven quality prediction methods are mainly built from static models using clean data with a slow sampling rate, leaving the process dynamics unused. To make full use of dynamic process data collected at a fast sampling rate, this paper proposes a novel deep learning-based robust dual-rate dynamic data modeling method for quality prediction of dynamic nonlinear processes. A new dynamic data denoising generative adversarial imputation network (DDGAIN) is first proposed for the missing value imputation among the dynamic process data. Then, a new hint convolutional neural network (HCNN) is established for dual-rate data based quality prediction. The proposed HCNN incorporates the information hint mechanism of channel expansion into the convolutional neural network to extract the dynamic features with definitive time and variable information. Finally, the proposed method is verified using the DOW distillation process dataset and Beijing multi-site air quality dataset.
AB - Traditional data-driven quality prediction methods are mainly built from static models using clean data with a slow sampling rate, leaving the process dynamics unused. To make full use of dynamic process data collected at a fast sampling rate, this paper proposes a novel deep learning-based robust dual-rate dynamic data modeling method for quality prediction of dynamic nonlinear processes. A new dynamic data denoising generative adversarial imputation network (DDGAIN) is first proposed for the missing value imputation among the dynamic process data. Then, a new hint convolutional neural network (HCNN) is established for dual-rate data based quality prediction. The proposed HCNN incorporates the information hint mechanism of channel expansion into the convolutional neural network to extract the dynamic features with definitive time and variable information. Finally, the proposed method is verified using the DOW distillation process dataset and Beijing multi-site air quality dataset.
KW - Data-driven quality prediction
KW - dual-rate data modeling
KW - dynamic data denoising generative adversarial imputation network
KW - dynamic process
UR - http://www.scopus.com/inward/record.url?scp=85162914965&partnerID=8YFLogxK
U2 - 10.1109/TII.2023.3275700
DO - 10.1109/TII.2023.3275700
M3 - Journal article
SN - 1941-0050
SP - 1
EP - 11
JO - IEEE Transactions on Industrial Informatics
JF - IEEE Transactions on Industrial Informatics
ER -