A typical design procedure for model predictive control or control
performance monitoring consists of:
- identification of a parametric or nonparametric model;
- derivation of the output predictor from the model;
- design of the control law or calculation of performance indices
according to the predictor.
Both design problems need an explicit model form and both require this
three-step design procedure. Can this design procedure be simplified?
Can an explicit model be avoided? With these questions in mind the work
presented in this book forms a new design paradigm that eliminates the
first and second step of the above design procedure. The subjects
treated include:
- closed-loop subspace identification;
- predictive control design;
- multivariate control performance assessment.
The approach presented in this book can be considered to be
"data-driven" in the sense that no traditional parametric models are
used; hence, the intermediate subspace matrices, which are obtained
directly from the process data and otherwise identified as a first step
in the subspace identification methods, are used directly for the
designs. Without using an explicit model, the design procedure is
greatly simplified and the modelling error caused by parameterization is
eliminated.