Converging Column Model

Two Liquid phases were found in column ATM-100 on stags 27 Vlaln TSi 

Irtftinnl Eiilerral Stream Inlet Slaae Transfer Basis Split  

IntetnalStieapfi External Stream Outlet Staaa Tvds Ttanslar Basis  

LIGHT NAPHTH LIGHT NAPH Condensei L P-H Flash  

---

Figure 4.4. Converged recycle results for the shortcut column model.

After we have completed the steam specifications, the column model will run automatically and quickly converge to the solution. We may receive warnings about a potential aqueous phase in the Light Naphtha stream. We will ignore these warnings until we complete building the entire column model. 

Figure 3.21 represents the process flow diagram of the rigorous model built by Aspen HYSYS. The model includes makeup gas streams and one column model to represent the VDU. In addition, we also add a heat flow to the feed stage to ensure that the flash zone temperature matches the plant measurement (Figure 3.22). The specifications of rigorous VDU model are similar to ADU model circulation rate and temperature change of each pumparound stream, flow rates of all liquid products except for one, top temperature (condenser temperature for ADU), and overflash rate/flash zone temperature [1]. By using the results of the simplified model first, we are able to apply these specifications to converge the VDU model with little effort...

Figure 3.23 represents the pressure, temperature and internal flow profiles obtained from the results of the simplified model. These profiles are used as initial estimates to enhance the convergence of the rigorous VDU model. We use all major liquid products except for VR, most of the circulation rates and temperature changes of pumparound streams, and flash zone temperature to specify the column model as shown in Figure 3.24. Figures 3.25 to 3.27 show the predictions of the rigorous VDU model for column temperature profile, D1160 curve of VGO and product yields. The results demonstrate that the two-step approach of model development generates accurate predictions on key operation and production variables of VDU.

Step 10 After the column model is converged, click record in data recorder . 

Once we converge the column models using the basic initialization specificahons, we change (especially for the main fractionator) to specificahons based on cut point and stage temperature. Kaes [51] describes a similar process. We summarize the inihal and final specificahons in Table 4.8. 

At this point in the inside-out method, the revised column profiles of temperature and phase compositions are used in the outer loop with the complex SRK thermodynamic models to compute updates of the approximate K and H constants. Then only one inner-loop iteration is required to obtain satisfactory convergence of the energy equations. The K and H constants are again updated in the outer loop. After one inner-loop iteration, the approximate K and H constants are found to be sufficiently close to the SRK values that overall convergence is achieved. Thus, a total of only 3 outer-loop iterations and 4 inner-loop iterations are required. 

Relaxation methods are not competitive with the steady-state methods in the use of computer time, because of slow convergence. However, because they model the actual operation of the column, convergence should be achieved for all practical problems. The method has the potential of development for the study of the transient behaviour of column designs, and for the analysis and design of batch distillation columns. 

In part I, Timm and Rachow Q) describe an algorithm for interpretation of chromatograms for imperfect resolution. The instrument was one of low plate counts, and yet population density di.,. ributions consistent with theoretical, kinetic models were achieved (, ) Research, using high plate count columns, shows that convergent distributions are achieved and that results are not a function of instrument resolution. Linear polystyrene resins had a polydispersity in the interval 1.5 M /M 2.0. 

Example 4.32 Column grand composite curves in methanol plant Table 4.16 describes the existing base case operations for columns 1 and 2 of the methanol plant obtained from the converged simulations using the RKS equation of state to estimate the vapor properties. The activity coefficient model, NRTL, and Henry components method are used for predicting the equilibrium and liquid properties. 

It is advisable to use the plant set points in building a model. For example, it is possible that a column simulation might involve the specification of the reflux ratio and bottoms flow rate because such specifications are (relatively) easy to converge. It is quite likely that the column may be controlled by using the temperature at some specific location (e.g., the temperature of tray 48). This specification should be used in building the model. In this case we do not know the set points used for controlling the column, but we do have sufficient information to allow us to compute the reflux ratio from plant flow data R = 11.6). Finally, the bottom product flow is specified as equal to the adjusted value reported above (17,999 kg/h). 

Solving the NEQ Model Equations In general, a nonequilibrium model of a column has many more equations than does an equivalent equilibrium-stage model. Nevertheless, we use may essentially the same computational approaches to solve the nonequilibrium model equations simultaneous convergence (Krishnamurthy and Taylor, op. cit.) and continuation methods [Powers et al., Comput Chem. Engng., 12, 1229 (1988)]. Convergence of a nonequilibrium model is likely to be slower than that of the equilibrium model because of the greater... 

In this particular case the converged composition and temperature profiles have the same shape as those obtained with the equilibrium-stage model (with specified efficiency) and, therefore, are not shown. The reason for the similarity is that, as noted above, this is basically a binary separation of very similar compounds. The important point here is that, unlike the equilibrium-stage model simulations, the nonequilibrium model predicted how the column would perform no parameters were adjusted to provide a better jit to the plant data. That is not to say, of course, that NEQ models cannot be used to fit plant data. In principle, the mass-transfer coefficients and interfacial area (or parameters in the equations used to estimate them) can be tuned to help the model better fit plant data. 

---