Performance assessment of process with uncertainties based on minimum variance

As for performance assessment based on MV (minimum variance) benchmark for MIMO systems, the knowledge of Markov parameter matrices is at least required for calculation of MV benchmark. However, in some occasions such as plant uncertainty, it is difficult to obtain the exact Markov parameter matrices. This work is to find the solution for MV-based performance assessment of MIMO systems in the presence of plant uncertainty. In the proposed technique, it shows that only delay matrix is required and the MV benchmark is obtained without any knowledge of interact matrix or Markov parameter matrices when conditions regarding the delay matrix is satisfied. Effectiveness of the proposed technique is demonstrated by Shell heavy oil fractionator process whose gain is uncertain.