869 ISSN 1999-8716 Printed in Iraq Second Engineering Scientific Conference College of Engineering –University of Diyala 16-17 December. 2015, pp. 869-884 MULTI-OBJECTIVE OPTIMIZATION OF SYNDIOTACTIC POLYMERIZATION OF STYRENE USING GENETIC ALGORITHM TECHNIQUE S. R. Sultan, Z. M. Shakoor, A. M. Hameed Chemical Engineering Departments, University of Technology, Iraq ABSTRACT: - The optimal control policies for the syndiotactic polymerization of styrene over silica supported metallocene catalyst, have been determined using a multiobjective optimization technique. Kinetics model (KM) and genetic algorithms (GA) were tested as tools for modeling and optimization of syndiotactic polystyrene (sPS) synthesis process. The dependence between the main parameters of the process and working conditions were modeled by using KM. To verify the KM, syndiotactic polymerization of styrene over silica supported metallocene catalyst was conducted. The validation results show that the KM predicts best polymerization reactor performance with an average absolute error less than 15%. The KM is then included into an optimizing control scheme, which uses a genetic algorithm solving technique and a multiobjective function in a scalar form. Genetic algorithms based methodology provides accurate results, computing optimal values of decision variables, which lead to the maximum rate of polymerization and the desired value for molecular weight. The validation results in these optimum values are valid and the average absolute error less than 5 % of all responses. Keywords: Multiobjective optimization, Genetic algorithms, Kinetics model, Polystyrene, Syndiotactic Polymerization. 1- INTRODUCTION Syndiotactic polystyrene (sPS) is a new polymeric material of industrial relevance, the high crystallization rate and the high melting point (270°C), make this polymer a crystalline engineering thermoplastic material with potential applications [1]. Syndiotactic polystyrene was first synthesized by Ishihara [2], using a soluble titanocene compound, and activated by methylalumoxane (MAO). Several styrene polymerization were carried out with supported metollocene catalyst, prepared by reaction of silica gel with MAO and then with metallocene catalyst [3, 4]. The optimization approach can have a significant effect on the polymer manufacturing process and economics. Polymer production facilities focus attention on improving the product quality and cost reductions [5]. In general, polymerization process optimization is naturally multi-objective, since it normally has several objectives that are often conflicting and non-measurable, which must be adjusting simultaneously. Therefore, solving such a problem cannot be devoid of difficulties, starting with the objective function formulating. It then proceeds with the choice of working procedure and the result selection from several options [6]. Multi-objective optimization can be defined as the problem of finding a vector of decision variables, which satisfies the constraints and optimizes a vector function whose elements represent the objective functions. In such cases, instead of obtaining a unique optimal solution, a set of equally good optimal solutions is usually obtained. There are referred to as Pareto sets [7]. A decision maker can choose any one of these non-dominant Diyala Journal of Engineering Sciences