Journal article · 2026 · Open access

Causal Graph Spatial--Temporal Autoencoder for Reliable and Interpretable Process Monitoring

Xiangrui Zhang, Chunyue Song, Wei Dai, Zheng Zhang, Kaihua Gao, Furong Gao

IEEE Transactions on Neural Networks and Learning Systems, 37(8), 3887–3899 · Published 4 February 2026 · DOI 10.1109/tnnls.2026.3656591

Open PDFView at the publisherCite this paper

In brief

“Causal Graph Spatial--Temporal Autoencoder for Reliable and Interpretable Process Monitoring” is a 2026 journal article by Xiangrui Zhang and 5 co-authors, published in IEEE Transactions on Neural Networks and Learning Systems. OpenAlex counts 31 citations as of 17 November 2026, which puts it in the top 1% of works published in 2026 by citations. It is open access under a CC BY licence: the full text is free to read at arXiv (Cornell University).

Citations
31
Most cited in
2026 (31)
Field-weighted impact
32.22
References
39
Authors
6

Abstract

To improve the reliability and interpretability of industrial process monitoring, this article proposes a causal graph spatial-temporal autoencoder (CGSTAE). The network architecture of CGSTAE combines two components: a correlation graph structure learning module based on spatial self-attention mechanism (SSAM) and a spatial-temporal encoder-decoder module utilizing graph convolutional long short-term memory (GCLSTM). The SSAM learns correlation graphs by capturing dynamic relationships between variables, while a novel three-step causal graph structure learning algorithm is introduced to derive a causal graph from these correlation graphs. The algorithm leverages a reverse perspective of causal invariance principle to uncover the invariant causal graph from varying correlations. The spatial-temporal encoder-decoder, built with GCLSTM units, reconstructs time series process data within a sequence-to-sequence framework. The proposed CGSTAE enables effective process monitoring and fault detection through two statistics in the feature space and residual space. Finally, we validate the effectiveness of CGSTAE in process monitoring through the Tennessee Eastman process (TEP) and a real-world air separation process (ASP).

Cite this paper

Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.

APA 7

Zhang, X., Song, C., Dai, W., Zhang, Z., Gao, K., & Gao, F. (2026). Causal Graph Spatial--Temporal Autoencoder for Reliable and Interpretable Process Monitoring. IEEE Transactions on Neural Networks and Learning Systems, 37(8), 3887–3899. https://doi.org/10.1109/tnnls.2026.3656591

MLA 9

Zhang, Xiangrui, et al. “Causal Graph Spatial--Temporal Autoencoder for Reliable and Interpretable Process Monitoring.” IEEE Transactions on Neural Networks and Learning Systems, vol. 37, no. 8, 2026, pp. 3887–3899. https://doi.org/10.1109/tnnls.2026.3656591.

Chicago

Zhang, Xiangrui, Chunyue Song, Wei Dai, Zheng Zhang, Kaihua Gao, and Furong Gao. 2026. “Causal Graph Spatial--Temporal Autoencoder for Reliable and Interpretable Process Monitoring.” IEEE Transactions on Neural Networks and Learning Systems 37 (8): 3887–3899. https://doi.org/10.1109/tnnls.2026.3656591.

Harvard

Zhang, X. et al. (2026) ‘Causal Graph Spatial--Temporal Autoencoder for Reliable and Interpretable Process Monitoring’, IEEE Transactions on Neural Networks and Learning Systems, 37(8), pp. 3887–3899. doi:10.1109/tnnls.2026.3656591.

IEEE

X. Zhang, C. Song, W. Dai, Z. Zhang, K. Gao, and F. Gao, “Causal Graph Spatial--Temporal Autoencoder for Reliable and Interpretable Process Monitoring,” IEEE Transactions on Neural Networks and Learning Systems, vol. 37, no. 8, pp. 3887–3899, 2026, doi: 10.1109/tnnls.2026.3656591.

BibTeX

@article{zhang2026causal,
  title     = {{Causal Graph Spatial--Temporal Autoencoder for Reliable and Interpretable Process Monitoring}},
  author    = {Zhang, Xiangrui and Song, Chunyue and Dai, Wei and Zhang, Zheng and Gao, Kaihua and Gao, Furong},
  journal   = {IEEE Transactions on Neural Networks and Learning Systems},
  year      = {2026},
  volume    = {37},
  number    = {8},
  pages     = {3887--3899},
  doi       = {10.1109/tnnls.2026.3656591},
  url       = {https://doi.org/10.1109/tnnls.2026.3656591}
}

Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)

Topic and keywords

Physical Sciences › Engineering › Control and Systems Engineering › Fault Detection and Control Systems

  • industrial process monitoring
  • fault detection
  • spatial self-attention
  • causal invariance
  • correlation graph
  • Tennessee Eastman process

Work with this paper

Most-cited papers in Engineering

All highly cited Engineering papers

Questions about this paper

How do I cite this paper in APA style?

Zhang, X., Song, C., Dai, W., Zhang, Z., Gao, K., & Gao, F. (2026). Causal Graph Spatial--Temporal Autoencoder for Reliable and Interpretable Process Monitoring. IEEE Transactions on Neural Networks and Learning Systems, 37(8), 3887–3899. https://doi.org/10.1109/tnnls.2026.3656591

How many times has this paper been cited?

31 times, according to OpenAlex on 17 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.

Is this paper free to read?

Yes. It is open access at arXiv (Cornell University): sign in and open the PDF from this page.

Who wrote this paper?

Xiangrui Zhang, Chunyue Song, Wei Dai, Zheng Zhang, Kaihua Gao, Furong Gao.

What is the DOI of this paper?

10.1109/tnnls.2026.3656591, permanently at https://doi.org/10.1109/tnnls.2026.3656591. Sign in and open it from this page.

Bibliographic data from OpenAlex (CC0), last checked 17 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.