Identification
Predictive Control : Process Modeling

The identification and calibration of the model from experimental data begins with the acquisition and processing of input/output signals from the application of a test protocol to the process being modeled. These signals are acquired from the control system (PLC...) in sampled discrete form. They are first pre-processed and filtered. The actual identification operation can then begin. All this pre-processing has a major influence on the quality of the resulting model.

3.1.1 Sampling