TY - JOUR
T1 - Multiobjective optimization of industrial naphtha cracker for production of ethylene and propylene
AU - Gujarathi, Ashish M.
AU - Motagamwala, Ali Hussain
AU - Babu, B. V.
PY - 2013/7/3
Y1 - 2013/7/3
N2 - Naphtha pyrolysis is one of the important routes for simultaneous production of ethylene and propylene. With recent increase in demand of both ethylene and propylene, understanding of naphtha pyrolysis becomes important for producing with increasing yield of these valuable products. Simultaneous maximization of yield of these two products is mathematically formulated as a multiobjective optimization (MOO) problem. Improved selected scheme is incorporated in the existing multiobjective differential evolution (MODE) algorithm and a new evolutionary algorithm is proposed. Both the evolutionary algorithms [i.e., MODE III and MODE-III with improved selection scheme (MODE III-ISS)] are used for MOO of industrial naphtha cracker unit. Two objectives (maximization of ethylene yield and propylene yield) and decision variables [pressure of the reactor tube (P), temperature of the reactor (T), initial flow rate of naphtha (F 0), and steam to naphtha ratio (SOR)] are considered for MOO study. MODE III and MODE III-ISS algorithms results are compared and presented, which clearly shows that the proposed MODE III-ISS algorithm possesses certain advantages over the MODE III algorithm (such as number of successful selections and percentage convergence with respect to initial number of population points, the quality of the obtained nondominated [ND] solutions).
AB - Naphtha pyrolysis is one of the important routes for simultaneous production of ethylene and propylene. With recent increase in demand of both ethylene and propylene, understanding of naphtha pyrolysis becomes important for producing with increasing yield of these valuable products. Simultaneous maximization of yield of these two products is mathematically formulated as a multiobjective optimization (MOO) problem. Improved selected scheme is incorporated in the existing multiobjective differential evolution (MODE) algorithm and a new evolutionary algorithm is proposed. Both the evolutionary algorithms [i.e., MODE III and MODE-III with improved selection scheme (MODE III-ISS)] are used for MOO of industrial naphtha cracker unit. Two objectives (maximization of ethylene yield and propylene yield) and decision variables [pressure of the reactor tube (P), temperature of the reactor (T), initial flow rate of naphtha (F 0), and steam to naphtha ratio (SOR)] are considered for MOO study. MODE III and MODE III-ISS algorithms results are compared and presented, which clearly shows that the proposed MODE III-ISS algorithm possesses certain advantages over the MODE III algorithm (such as number of successful selections and percentage convergence with respect to initial number of population points, the quality of the obtained nondominated [ND] solutions).
KW - Differential evolution
KW - Evolutionary algorithms
KW - Mathematical models
KW - Multiobjective optimization
KW - Naphtha
UR - http://www.scopus.com/inward/record.url?scp=84880459161&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84880459161&partnerID=8YFLogxK
U2 - 10.1080/10426914.2012.746706
DO - 10.1080/10426914.2012.746706
M3 - Article
AN - SCOPUS:84880459161
SN - 1042-6914
VL - 28
SP - 803
EP - 810
JO - Materials and Manufacturing Processes
JF - Materials and Manufacturing Processes
IS - 7
ER -