The flowsheet shown in Figure 10.15 does not show the plumbing required to run a realistic pressure-driven dynamic simulation. The key feature is that the pressure in the stripper must be greater than that in the main column so that vapor can flow from the top of the stripper back to the main column. Therefore, in the simulation, a pump and a control valve are placed in the liquid sidestream. A control valve is also placed in the stripper overhead vapor line. All this plumbing is shown in Figure 10.16. In a real physical setup it is usually possible to use elevation differences to provide the necessary differential pressure driving force to get the liquid to flow from the main column into the stripper at a higher pressure and avoid the use of a pump.
The usual sizing calculations are performed for both the main column and the stripper (10-min liquid holdups in column bases and reflux drum). The flowsheet is pressure-checked, and the file is exported to Aspen Dynamics.
A control structure is developed for this more complex system. The various loops are described below. The key issues are how to manipulate the sidestream and how to maintain the compositions of the three product streams.
1. Feed is flow controlled.
2. With a reflux ratio of 5, the reflux drum level is controlled by manipulating reflux flowrate.
3. The temperature on stage 4 of the main column is controlled by manipulating distillate.
4. Base level is controlled by manipulating bottoms flowrate in both columns.
5. Pressure in the main column is controlled by manipulating condenser heat removal.
6. Pressure in the stripper is controlled by manipulating the valve in the vapor line V22.
7. The temperature on stage 51 in the main column is controlled by manipulating reboiler heat input to the main column.
8. The temperature on stage 5 in the stripper is controlled by manipulating stripper reboiler heat input.
9. The sidestream flowrate is ratioed to reflux flowrate.
This last loop is the most important feature of this control structure. It permits changes in the sidestream as disturbances enter the system.
Figure 10.17 Temperature profiles.
Locations of the temperature control trays are selected by looking at the temperature profiles in the two columns, shown in Figure 10.17. The three temperature controllers (TC4 and TC51 in the main column and TC5 in the stripper) have 1-min deadtimes and 100 K temperature transmitter spans. They are tuned individually by running a relay -feedback test with the other two controllers on manual and using the Tyreus -Luyben settings. The controller tuning constants are given in Table 10.1.
Note that the TC4 temperature control loop is "nested" inside the reflux drum level controller since a change in distillate flowrate has no direct effect on stage 4 temperature. The tuning of the reflux drum level controller (LC12) affects the tuning of the TC4 temperature controller, as Table 10.1 shows. Performance is improved by tightening up on the level controller. This is illustrated in Figure 10.18 for a 20% increase in feed flowrate.
The responses for feed composition disturbances are shown in Figure 10.19. In Figure 10.19a the feed composition is changed from 35 mol% DME to either 40 or 30 mol% DME (with a corresponding change in MeOH). In Figure 10.19b the feed
TABLE 10.1 Stripper Temperature Controller Parameters
TC4 with LC12 Gain = 2
TC4 with LC12 Gain = 5
Was this article helpful?