Excess Emission Sources Identification via Sparse Monitoring Networks

Kuang Cheng,Xiangyu Zhao,Wang Zhou,Yi Cao,Shuanghua Yang
DOI: https://doi.org/10.1109/tii.2023.3240560
IF: 12.3
2023-01-01
IEEE Transactions on Industrial Informatics
Abstract:Abnormality detection and identification is of great concern for complex processes monitoring. In chemical industrial parks (CIPs), the primary task is to identify abnormal (excess) sources when unexpected excess emissions occur. However, it is an ill-posed problem to detect, especially to identify them in dense industrial areas where a vast number of emission sources are concentrated in limited space, challenging the relatively sparse wireless sensor networks (WSNs). Meanwhile, barely detecting the existence rather than identifying excess sources can hardly meet the requirements of fine management in CIPs. In this paper, a QR decomposition-based method for excess sources identification (QR-ESI) has been proposed. By introducing equivalent sources (ES) as a substitute for real sources (RS), the semi-independent relation between ES and RS has been developed. By monitoring ES, excess sources can be inferred progressively through the iteration of logical judgments. The performance was evaluated with simulated data and then validated and compared with a state-of-the-art method in a case study in a real-world chemical industrial park.
automation & control systems,computer science, interdisciplinary applications,engineering, industrial
What problem does this paper attempt to address?