Causal intervention graph neural network for fault diagnosis of complex industrial processes

Ruonan Liu, Quanhu Zhang, Di Lin, Weidong Zhang, Steven X. Ding

Abstract: Highlights•Formulates GNN-based fault diagnosis as a graph classification task.•Analyzes the causality in GNN-based fault diagnosis process based on causal theory.•Constructs automatically sensor signals into graph data using an attention mechanism.•Designs an instrumental variable for causal intervention on graphs.