Attainability condition, CSTR

A key difference between the complement method and the LP method is that the latter requires the solution of a linear program, whereas the former is a direct application of the CSTR attainability condition. In the LP approach, all points on the AR boundary are computed simultaneously—via the solution of a large linear program—in a single calculation step. In order for this result to be achieved, the candidate region boundary points must be expressed in terms of all other boundary points in space using a superstructure formulation, which is termed the total connectivity model. 

Over 25 years ago the coking factor of the radiant coil was empirically correlated to operating conditions (48). It has been assumed that the mass transfer of coke precursors from the bulk of the gas to the walls was controlling the rate of deposition (39). Kinetic models (24,49,50) were developed based on the chemical reaction at the wall as a controlling step. Bench-scale data (51—53) appear to indicate that a chemical reaction controls. However, flow regimes of bench-scale reactors are so different from the commercial furnaces that scale-up of bench-scale results caimot be confidently appHed to commercial furnaces. For example. Figure 3 shows the coke deposited on a controlled cylindrical specimen in a continuous stirred tank reactor (CSTR) and the rate of coke deposition. The deposition rate decreases with time and attains a pseudo steady value. Though this is achieved in a matter of rninutes in bench-scale reactors, it takes a few days in a commercial furnace. 

There are several control problems in chemical reactors. One of the most commonly studied is the temperature stabilization in exothermic monomolec-ular irreversible reaction A B in a cooled continuous-stirred tank reactor, CSTR. Main theoretical questions in control of chemical reactors address the design of control functions such that, for instance (i) feedback compensates the nonlinear nature of the chemical process to induce linear stable behavior (ii) stabilization is attained in spite of constrains in input control (e.g., bounded control or anti-reset windup) (iii) temperature is regulated in spite of uncertain kinetic model (parametric or kinetics type) or (iv) stabilization is achieved in presence of recycle streams. In addition, reactor stabilization should be achieved for set of physically realizable initial conditions, (i.e., global... 

Stability would occur if there are initial conditions in the reactor Cas T (subscript S for steady state) from which the system will evolve into each of these steady states. Also, if we start the reactor at exactly these steacfy states, the system will remain at that state for long times if they are stable. We will look at the first case later in connection with transients in the nonisothermal CSTR. Here we examine stability by asking if a steady state, once attained, will persist. 

In other situations where the emulsifier feedrate to the reactor is sufiicient to produce a steady-state concentration in the CSTR that is above the CMC, one observes damped oscillations upon start-up followed by an eventual attainment of a steady-state, nonosdllatory condition (Kiparissides et I980a,b). 

Bench-scale kinetic experiments can be conducted in batch, continuous stirred-tank (CSTR), tubular plug-flow, or differential reactors. The last of these can be operated with once-through flow or recycle. Advantages and disadvantages of the various types are discussed. In particular Batch reactors are inexpensive, but require attention to rapid attainment of reaction conditions at start CSTRs are excellent for gas-liquid, but less so for gas-phase reactions tubular reactors are especially suited for reactions of heterogeneous catalysis and differential reactors operated "once through" are best for measurement of initial rates. 

The third condition can help to extend a candidate region that might fulfil the first two conditions. The stationary point with mixing is in fact a CSTR. If such point can be found outside the candidate region, this means that the Attainable Region has not been yet found, and an extension is possible. This condition can be satisfied if the system feed and maximum rate reactors (both absolute and relative) have been considered. 

However the region at the left of the point P is not convex. But we may continue from the point P with a PFR, as indicated in Fig. 8.27B. The computation is done by simply Integrating the differential equations of a PFR, this time input concentrations supplied by the exit of the CSTR. The new augmented region is convex. This time all three conditions are fulfilled. No other mixed reactors can be found above the boundary that could give a higher amount of B. This is the final Attainable Region. 

Figure 6.1 illustrates the fact that for various ranges of kinetic and reactor parameters it is possible for the mass and energy conservation relations for a CSTR to be in stable balance at more than one condition. This may imply that there are other balance conditions that are unstable the point needs to be examined. Which of the stable balances is attained in actual operation may be dependent on the details of startup procedure, for example, which are not subject to the control of the designer. Thus, it is important to investigate reactor stability using unsteady-state rather than steady-state models. 

All three effluent concentrations satisfy the CSTR equation, and thus all three are valid solutions. The actual concentration attained if a CSTR were to be operated with the same conditions would depend on other factors, such as the concentration in the CSTR at start-up, or whether one of the calculated solutions are stable operating conditions. ... 

If a point C in the complement region does not satisfy the CSTR condition, then it simply means that it is not achievable, by the CSTR condition, at the current iteration. However, C might still be attainable at a later point ... 

We also described how concrete equations for critical DSR and CSTRs may be computed. These expressions are complicated to compute analytically, which are derived from geometric controllability arguments developed by Feinberg (2000a, 2000b). These conditions are intricate, and thus it is often not possible to compute analytic solutions to the equations that describe critical reactors. For three-dimensional systems, a shortcut method involving the vDelR condition may be used to find critical a policies. Irrespective of the method used, the conditions for critical reactors are well defined, irrespective of the legitimacy of the kinetics studied, and thus these conditions must be enforced if we wish to attain points on the true AR boundary. 

NOTE There is more than one possible steady-state condition that may be attained by the non-isothermal CSTR. In other words,, P", and T may each assume a value that is different but dependent on the other two. Moreover, the ode solver ode 15s (for a stiff system of ODEs) was used instead of ode45, for the latter went in vain. Consequently, if the user is stuck with the first-to-try method then he or she ought to use other MATLAB ode solvers. Please refer to Table 7.3 in Sec. 7.3 to see when to use one method over others. 

---