A novel controller performance index using a local process model factorization

Souza, D.L., Oliveira-Lopes, L.C.


Abstract:

This work introduces a measure for control performance assessment by using process model information and input-output plant data. Knowledge of the invariants of the process and tools to factorize local linear models to estimate the generalized interactor matrix are required and a reference behavior is designed with the action of the current control implemented into a fictitious plant designed in such a way that the major feedback difficulties are taken out from the control problem. The performance index assessment (GIM) is illustrated to MIMO systems and a comparison with MVC based indexes is carried out. The results indicate greater consistency of the proposed methodology when compared to existing techniques. This fact allows to the practitioner to perform the implementation of the control assessment with easy. The paper also shows a comparison between different performance indexes for process control assessment.

Volume: 6 | Issue: 3

Pages: 8-23

Purchase this Article