Parameter Estimation of Dynamic Grade Transitions in a Polyethylene Plant

A polyethylene plant at Borealis AB is modelled in the Modelica language and considered for dynamic parameter estimation at grade transitions using plant measurement data. To solve the parameter estimation problem, the JModelica.org platform is used, offering tools for simulation with parameter sensitivies. The model contains several parameters, from which the best parameter sets have been chosen using a parameter selection algorithm. Cross validations of the data sets show promising results. The resulting model will be used for dynamic optimisations of grade changes.