Page 298 - Distillation theory
P. 298

P1: JPJ/FFX  P2: FCH/FFX  QC: FCH/FFX  T1: FCH
            0521820928c08  CB644-Petlyuk-v1                                                      June 11, 2004  20:20





                        272    Synthesis of Separation Flowsheets

                                    F 1,J = 0
                                    F 2,J = S 2,J,1
                                    F 3,J = min (S 3,J,K + F K−J+1,J + F 3−K,K+1 )              (8.6)
                                          K 3,J
                                   .............................................................
                                    F n,1 = min (S n,1,K + F K1 + F n−K, K+1 )
                                          K n,1
                                 The method of dynamic programming synthesizes the optimal sequence start-
                               ing from its end. Therefore, expenditures S I,J,K should be determined without
                               calculation of the previous part of the flowsheet. For this purpose, it is necessary
                               to determine the composition of feeding of column I,J,K. It can be done easily,
                               if it is accepted that each product of separation sequence i contains as impurity
                               components only adjacent components (i − 1) and (i + 1) (i.e., the set permissible
                               concentrations of impurity components):
                                     L    i−1
                                    η = η                                                      (8.7a)
                                     i    i
                                     H    i+1
                                    η = η                                                      (8.7b)
                                     i    i
                                             H
                                      L
                               Where η and η are set permissible concentrations of light and heavy impurity
                                      i      i
                               components in product i correspondingly, η i−1  and η i+1  concentrations of compo-
                                                                    i      i
                               nents (i − 1) and (i + 1) in product i correspondingly.
                                 At this assumption, the amount P i of the product i can be determined from
                               the system of linear equations of componentwise material balance that has three-
                               diagonal form:
                                               H
                                    f 1 = P 1 (1 − η ) + P 2 η L
                                               1       2
                                                      L
                                           H
                                                           H
                                    f 2 = P 1 η + P 2 (1 − η − η ) + P 3 η L                    (8.8)
                                                                   3
                                                           2
                                                      2
                                           1
                                   ....................................................
                                                          L
                                    f n = P n−1 η H  + P n (1 − η )
                                             n−1         n
                               After that, the feeding of column I, J can be determined:
                                     I,J
                                    f   = 0    (for i < J − 1)
                                     i
                                    f  I,J  = P J η L  (for i = J − 1)
                                     i       J
                                    f  I,J  = f J − P J−1 η H  (for i = J)
                                     i             J−1
                                    f i I,J  = f i  (for J + 1 ≤ i ≤ J + I−2)                   (8.9)
                                     I,J               L
                                    f   = f J+I−1 − P J+I η   (for i = J + I−1)
                                     i                 J+I
                                     I,J         H
                                    f   = P J+I−1 η      (for i = J + I)
                                     i           J+I−1
                                    f  I,J  = 0  (for i ≥ J + I + 1)
                                     i
                                 The above-described algorithm was used for the synthesis of separation flow-
                               sheet of mixture of hydrocarbon gases C 3 H 8 , i-C 4 H 10 , n-C 4 H 10 , i-C 5 H 12 , n-C 5 H 12 ,
                               i-C 6 H 14 , and n-C 6 H 14 (n = 7). At (n = 7), the number of alternative sequences is
                               132. Composition and flow rate of feed, permissible impurities in the products, and
                               flow rates of the products are given in Table 8.1. The comparison of alternative
                               sequences was made in accordance with the value of annual expenditures. The
                               calculation of the columns were executed in accordance with simplified method
                               Fenske–Underwood–Gilliland. Figure 8.2a shows the graph of dependence of total
   293   294   295   296   297   298   299   300   301   302   303