Fault Diagnosis of Regenerative Thermal Oxidizer System Via Dynamic Uncertain Causality Graph Integrated with Early Anomaly Detection

Shangbo Han,Yiyan Hua,Yangshu Lin,Longchao Yao,Zhongcheng Wang,Zhengjie Zheng,Jian Yang,Chunhui Zhao,Chenghang Zheng,Xiang Gao
DOI: https://doi.org/10.1016/j.psep.2023.09.058
2023-01-01
Abstract:Regenerative thermal oxidizers (RTO) serve as pivotal equipment for the treatment of volatile organic compounds (VOCs). However, a diversity of faults of the RTO system may threaten personnel safety and lead to unscheduled downtime. In this work, we develop a traceability inference method based on dynamic uncertain causality graph (DUCG) to identify root fault cause of RTO at early anomaly alarm. A general expert knowledge base for the fault diagnosis of RTO system is constructed. And the anomaly alarm thresholds are modified according to historical data with n-sigma rules to improve the diagnostic accuracy. Results on different cases show advantages in both triggering early anomaly alarms and avoiding invalid alarms, which is crucial to improve the fault inference accuracy. By applying our method to 35 fault cases of a real RTO for pharmaceutical VOCs treatment, the fault detection rate reaches 94.29%, and the fault inference accuracy is 82.86%. The two scores are 3% and 21% higher than those with the conventional DUCG (with fixed alarms), respectively. This remarkable improvement underlines the significance of our proposed method for RTO maintenance, and it is expected to assure the system’s operational stability and reliability.
What problem does this paper attempt to address?