Identification of State-space Models for Processes with Irregularly Sampled Outputs
Abstract
In many processes, variables which indicate product quality are irreg-ularly sampled. Often, the inter-sample behavior of these quality variables canbe inferred from manipulated variables (MV) and other process variables whichare measured frequently. When the quality variables are irregularly sampled,Maximum Likelihood Estimation (MLE) can be performed using the ExpectationMaximization (EM) approach. The initial model required for the EM algorithmcan be obtained using a realization-based subspace identification technique. Wedescribe such a model identification method and present its application on simu-lation and industrial case studies.
