Work

Graph autoencoder with causal relationship inference for fault detection and root cause identification in complex industrial process

Year: 2026

Type: article

Abstract: Modern industrial processes consist of multiple strongly coupled units, where malfunctions propagate among various process variables and units. This poses great challenges in identifying the root cause of process faults. This paper proposes a novel network called graph autoencoder with causal relationship inference (GACRI) to tackle this challenge. Firstly, a dual-view decoder graph autoencoder (i.e., feature-level and causal-level decoders) is proposed to consider temporal features and spatial causal relationships concurrently. Secondly, a global index based on the learned features and the reconstructed residual space is developed for fault detection. The variable contribution degree is analyzed to isolate the fault variables via the reconstruction loss. Finally, a causal discovery network is designed to predict the causal relationship among fault variables for root-cause identification. The testing results on three processes (i.e., a numerical process, the Tennessee Eastman process and a semiconductor process) demonstrate the superior performance of GACRI in process monitoring.

Source: Advanced Engineering Informatics

DOI: 10.1016/j.aei.2026.104408

Authors: Shijin Li, Jun Hao, Jianbo Yu

Institutions: Tongji University, Zhejiang College of Security Technology

Language: English

Topic: Fault Detection and Control Systems

Subfield: Control and Systems Engineering

Field: Engineering

Domain: Physical Sciences

SDG: Industry, innovation and infrastructure

Open Access status: hybrid

Estimated APC paid: $3,560