Adversarial-Causal Representation Learning Networks for Machine fault diagnosis under unseen conditions based on vibration and acoustic signals

Fei Wu, Zhuohang Xiang, Dengyu Xiao, Yaodong Hao, Yi Qin, Huayan Pu, Jun Luo

Abstract: Highlights•A novel spatial mask domain adversarial method is proposed to capture features.•By improving the loss function and narrowing the gap, deep information is mined.•A novel channel mask feature orthogonality method removes the redundant dependency.•Assessing similarity makes causal factors independent, avoiding spillover effects.•Extensive experiments validate ACRLN’s superiority in DG-based fault diagnosis tasks.