Two-level multiblock statistical monitoring for plant-wide processes

Due to the complexity of plant-wide processes, many of the current multivariate statistical process monitoring techniques are lacking in interpretation of the detected fault, and fault identification also becomes difficult. A new two-level multiblock independent component analysis and principal component analysis (MBICA-PCA) method is proposed in this paper. Different from the conventional method, the new approach can incorporate block information into the high level for global process monitoring. Through the new method, the process monitoring task can be greatly reduced and the interpretation for the process can be made more quickly. When a fault is detected, a two-step fault identification method is proposed. The responsible sub-block is first identified by contribution plots, which is followed by fault reconstruction in the corresponding sub-block for advanced fault identification. A case study of the Tennessee Eastman (TE) process evaluates the feasibility and efficiency of the proposed method.