Fault Diagnosis of Unseen Modes in Chemical Processes Based on Feature Weighted

Fault diagnosis is critical for ensuring production safety and product quality within the chemical industry. Existing fault diagnosis models that leverage deep neural networks are typically trained using samples from specific operating conditions, neglecting the domain generalization capabilities of these models. This leads to a significant decline in diagnostic performance in complex and variable scenarios. To address this issue, a Weighted-Based Domain-Specific Feature Removal Network (WBDSFRN) is proposed. WBDSFRN incorporates a weighted-based domain-specific feature removal module, which effectively discriminates between domain-invariant and domain-specific features during the training phase. In the testing phase, the model largely eliminates domain-specific features of the target domain to mitigate domain shift effects. Finally, experiments are conducted using the Tennessee-Eastman Process (TEP). The results demonstrate that WBDSFRN significantly outperforms existing approaches and exhibits robust diagnostic performance even under complex and variable operating conditions.