Joint-individual monitoring of large-scale chemical processes with multiple interconnected operation units incorporating multiset CCA

- A joint-individual monitoring scheme incorporating MCCA for large-scale processes is proposed.
- Both global information and local information are considered in the monitoring.
- The proposed monitoring scheme discriminates the characteristic of a detected fault.
- The monitoring performance is analyzed within the statistical framework of hypothesis testing.
- Case studies on a numerical example and the TE process indicate the efficiency.

Large-scale processes with multiple interconnected operation units have become popular, and monitoring such processes is imperative. A joint-individual monitoring scheme that incorporates multiset canonical correlation analysis (MCCA) for large-scale chemical processes with several interconnected operation units is proposed. First, MCCA is employed to extract the joint features throughout the entire process. Second, for each operation unit, the measurements are projected into a joint feature subspace and its orthogonal complement subspace that contains the individual features of the unit. Then, corresponding statistics are constructed to examine the joint and individual features simultaneously. The proposed joint-individual monitoring scheme considers the global information throughout the entire process and the local information of a local operation unit and therefore exhibits superior monitoring performance. The joint-individual monitoring scheme is applied on a numerical example and the Tennessee Eastman benchmark process. Monitoring results indicate the efficiency of the proposed monitoring scheme.