An efficient model reduction technique for the distillation columns is applied to account for the detail dynamics. This technique utilizes the orthogonal collocation and cubic spline method. Then the extended Kalman filter is applied to identify the model parameters and the feed composition from the measurements of the column. From the simulation, the model reduction technique can account for the detail dynamics of the rigorous distillation model and not only the model parameters, but also the feed composition can be identified by the recursive prediction error method.
ASJC Scopus subject areas
- General Chemical Engineering
- Computer Science Applications