Research On Modeling And Grade Transition Strategy In Polypropylene Process

Polypropylene (PP) possesses of many great merits, such as the light proportion, high resistance, corrosion stability, high diaphaneity, non-toxicity, good intensity, insulation and easy to be machining. Therefore, PP has been widely used in the light industry, chemical industry, chemical fiber, building materials, electrical household appliance, packaging and motordom. It is one of the backbone industries of our country.Grade transition is the operation which translates one process condition to another, huge economic loss is caused by frequent grade transition which brought by various demands made by market and competition. This thesis focuses on the modeling and optimization of grade transition in polypropylene process. The main research work and the achievements are listed as follows:1. The knowledge about PP was systematically introduced, and recent works on process control, modeling and identification of polypropylene process were given;2. The propylene polymerization and Spheripol process mechanism were introduced. Based on production data, the dynamic model of polypropylene process is established by RVM, which is sparser than SVM, and no parameter tuning is required to be involved. Simulation results shows that the model can predict the melt index to a certain extent;3. A modified differential evolution (MDE) algorithm with single population and preferential crossover is proposed for the dynamic and continuous optimization problems in chemical engineering. Single string and preferential crossover are employed to speed up evolution algorithm; Relocation to interior (RI) method which guarantees the search ability is introduced to handle constraint. A numerical study suggests that proposed modification can reduce function evaluation numbers and CPU time, enhance the search ability. Furthermore, the proposed MDE is used to solving dynamic optimization problems of batch reactor, simulation results illustrate the efficiency and effectiveness of the proposed methods;4. A modified differential evolution algorithm with dynamic scaling factor and adaptive mutation operator is proposed. Dynamic scaling factor is suitable for handling the constraint optimization problem. Mutation operator maintains population diversity to overcome the premature and enhance the probability of finding global optimum. Modified differential evolution algorithm is applied to optimal grade transition in polypropylene polymerization, the target function was built and calculated so as to get the optimal grade transition trajectories;5. Finally, the conclusions are presented, some problems to be solved and the possible directions on the development of polypropylene production were pointed out.