Data-driven process decomposition and robust online distributed modelling for large-scale processes

Abstract

With the increasing attention of networked control, system decomposition and distributed models show significant importance in the implementation of model-based control strategy. In this paper, a data-driven system decomposition and online distributed subsystem modelling algorithm was proposed for large-scale chemical processes. The key controlled variables are first partitioned by affinity propagation clustering algorithm into several clusters. Each cluster can be regarded as a subsystem. Then the inputs of each subsystem are selected by offline canonical correlation analysis between all process variables and its controlled variables. Process decomposition is then realised after the screening of input and output variables. When the system decomposition is finished, the online subsystem modelling can be carried out by recursively block-wise renewing the samples. The proposed algorithm was applied in the Tennessee Eastman process and the validity was verified.

Global Coordinated Distributed Predictive Control Algorithm Based on System Dynamic Decomposition Index

“…A data-driven system decomposition method has been proposed in [25] where process decomposition is realized by determining the input and output variables of the subsystem from the perspective of control. Canonical correlation analysis(CCA) algorithm is adopted to measure the static correlation between input and output variables and finally determine the input variables of a subsystem.…”

“…Hence, besides as the basis of system decomposition, these data analysis indices can also be utilized in the coordination strategy of DMPC algorithm. In the former work of our group, a system decomposition methodology based on AP-CCA has been studied [25]. In that algorithm static correlation between the process variables is used as the decomposition index.…”