Calibration of Physical Models with Process Data using FIR Filtering

摘要

Automatic calibration of physical plant models in the context of monitoring and control of industrial processes is considered. A structure integrating a physical model and estimated FIR filters is proposed. In addition, a finite state FIR structure is proposed to complement the calibrated physical model with a data-driven mapping. The approach is illustrated in simulations using the van der Vusse CSTR benchmark.