Dynamic analysis and system identification of a schiebel extraction column

A rigorous model for dynamic simulation of extraction process is developed. The simulation model is shown to be accurate for prediction of process behaviour under different operating conditions. Dynamic analysis is carried out on the simulation model to understand the process transient behaviour under different conditions. Step testing is then applied on the rigorous simulation model to generate input-output response data, which are then used for multivariable system identification in order to derive simple reduced-order linear models that can adequately capture the process dynamics. The identified model predictions are found to be in a good agreement with the rigorous ones. Also, these models are shown to be useful in control system design.