Control Structure Selection
Possible candidates for manipulated variables are essentially the same as in non-RD, including reflux, distillate flow rate or reflux ratio at the top of the column and heating rate, bottoms flow rate, or reboil ratio at the bottom of the column. In addition, dosing of the reactants can be an interesting choice in RD, if this is compatible with the upstream processing of the plant. Possible candidates for measured variables are either product compositions or column temperatures. However, online measurement of concentrations is usually slow, expensive, and often not very reliable. Therefore inferential control schemes are preferred, where the product compositions are inferred from temperature measurements. However, the relationship between product compositions and column temperatures is frequently non-unique in RD [26, 98] and this can lead to severe problems as will be illustrat-
ed in a first case study. This case study is inspired by the paper of Roat et al.  and is concerned with a pilot plant column for producing methyl acetate.
A first control scheme proposed in  is shown in Fig. 10.26. In this scheme, product purities of methyl acetate (MeAC) and water (H20) are inferred from temperatures on trays 3 and 12, respectively, and the feed rates of methanol (MeOH) and acetic acid (AcH) are used as manipulated variables. For this configuration, three different temperature profiles exist with identical temperature values at the sensor locations but different feed rates and completely different product compositions. The solid line in Fig. 10.26 represents the desired temperature profile with high conversion. This situation corresponds to input multiplicity as introduced at the beginning of section 10.2 on multiplicity and oscillations. Here, the same set of output variables (temperatures) is produced by (three) different sets of input variables (feed rates). Because the steady state values of the output variables are fixed by the given setpoint of the controllers, this input multiplicity will lead to steady state multiplicity of the closed loop system as illustrated in Fig. 10.27.
Fig. 10.27 shows the time plots of the input variables (feed rates), the output variables (temperatures on trays 3 and 12) and the product purities of methyl acetate in the distillate and water in the bottoms for two different pulse disturbances of the heating rate. Initially the column is at the desired steady state with high product purity. After a small disturbance, indicated by the dashed line, the column returns
Fig. 10.27 Dynamic transient behavior for control scheme 1 in Fig. 10.26 after a small (dashed line) and a large (solid line) pulse disturbance of the heating rate to the desired steady state. After a large disturbance, indicated by the solid line, the column suffers an undesired transition to a different steady state with low product purity. In both cases the controlled variables (temperatures on trays 3 and 12) are finally forced to their given setpoints by the controllers.
Because input multiplicity depends on the choice of the output variables, it can be avoided by selecting a suitable control configuration. In the present case, this problem is easily solved by taking a third temperature measurement into account as proposed in  and illustrated in Fig. 10.28. The difference between the two temperature measurements in the reaction zone is an indirect measure for the conversion of the chemical reaction and therefore leads to a unique steady state with good control characteristics. Similar control schemes, with direct and indirect inference of conversion were proposed in [99, 100, 105] for etherification processes.
Was this article helpful?