A semi-supervised temporal modeling strategy integrating VAE and Wasserstein GAN under sparse sampling constraints

Time series network models are widely applied in process industries for soft sensing, fault monitoring, and real-time optimization, serving as a powerful tool to enhance the safety and efficiency of industrial production. Typically, time series networks require labeled data for supervised learning....

Full description

Saved in:
Bibliographic Details
Published in:Journal of process control Vol. 152; p. 103497
Main Authors: Hu, Yujie, Xie, Changrui, Chen, Xi
Format: Journal Article
Language:English
Published: Elsevier Ltd 01.08.2025
Subjects:
LSTM
Semi-Supervised
TCN
Time series network
VAE
WGAN
LSTM
VAE
Semi-Supervised
WGAN
Time series network
TCN
ISSN:0959-1524
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Time series network models are widely applied in process industries for soft sensing, fault monitoring, and real-time optimization, serving as a powerful tool to enhance the safety and efficiency of industrial production. Typically, time series networks require labeled data for supervised learning. However, labeled data often exhibits sparse sampling characteristics in industrial settings, which limits the model's performance. To address this issue, a semi-supervised modeling strategy based on Variational Autoencoder (VAE) and Wasserstein Generative Adversarial Network (WGAN) is proposed in this paper. The strategy consists of three steps. First, for the labeled samples, process data and labeled data are used as input to train a supervised VAE model (SVAE). Upon completion of the training, the posterior distribution of the latent variable zS is obtained. Second, in all samples, only process data is used to train an unsupervised VAE model (UVAE) to extract the latent variable zU, and the WGAN discriminator is introduced to distinguish between "fake data" (zU) and "real data" (zS). Through adversarial learning between the UVAE and WGAN discriminator, the posterior distribution of zU is forced to approximate zS. Finally, the encoder of UVAE and the decoder of SVAE are combined to form a Semi-Supervised Variational Autoencoder model (SS-VAE), which extracts the latent variable zSS and the reconstructed labeled data from the decoder as inputs for the time series network. Long Short-Term Memory (LSTM) and Temporal Convolutional Network (TCN) are selected as two basic time series models, and their performance, both with and without the proposed semi-supervised approach, is compared to assess the effectiveness and robustness of the strategy. The improvements observed in two industrial case studies validate the efficiency of the proposed approach. •Combine VAE and WGAN to enhance modeling with sparse labeled data.•Align the latent space of the unsupervised VAE with the supervised VAE via adversarial training.•Use a semi-supervised VAE for enhanced deep feature extraction.•Allow different time series networks based on specific modeling tasks.•Achieve significant performance improvements in industrial case studies.
ISSN:0959-1524
DOI:10.1016/j.jprocont.2025.103497