Reconciliation of an Experimental Distillation Column

FIGURE 8 Time evolution of measured and reconciled values. 

For comparison purposes, plant data and reconciled data, both with and without bias treatment, over the whole operating cycle of the pyrolysis reactor for some variables are plotted in Fig. 8. This figure clearly shows that the measurements obtained from coil 2 are biased. There were not many differences between measured and reconciled values for the other variables. 

The overall heat transfer coefficient calculated using the joint parameter estimation and data reconciliation approach is shown in Fig. 9. It is evident from this figure that the overall heat transfer coefficient remains fairly constant throughout the whole operating cycle of the pyrolysis reactor. Near the end of the cycle, the heat transfer coefficient drops to a comparably low value, signifying that the reactor needs to be regenerated. 

The final case study consists of a well-instrumented experimental distillation column that has been interfaced to an industrial distributed control system (Nooraii 

FIGURE 9 Time evolution of the estimated heat transfer coefficient (from Weiss et al., 1996). 

---

Finally, a well-instrumented experimental distillation column that has been interfaced to an industrial distributed control system was used to show the implementation of the techniques described in previous chapters in an actual on-line framework, using industrial hardware. In this case, the usefulness of data reconciliation, prior to process modeling and optimization, was clearly demonstrated. 

---