Optimal Grade Transition for Polyethylene Reactors Based on Simultaneous Strategies and Trust Region Particle Swarm Optimization

详细信息    查看全文

• 作者：Wenxing Xu ; Zhiqiang Geng ; Qunxiong Zhu ; Xiangbai Gu
• 刊名：Industrial & Engineering Chemistry Research
• 出版年：2013
• 出版时间：March 6, 2013
• 年：2013
• 卷：52
• 期：9
• 页码：3363-3372
• 全文大小：477K
• 年卷期：v.52,no.9(March 6, 2013)
• ISSN：1520-5045

文摘

Various grades of polyethylene are required to be produced in the same industrial equipment by changing the operating conditions. However, transitions between different grades are rather slow and result in the production of a considerable amount of off-specification polymer. To improve grade transition is viewed here as a dynamic optimization problem, for which simultaneous strategies are especially beneficial. This paper first used a simultaneous strategy to fully discretize the state and control variables, converting the mathematical optimization model into a large-scale NLP problem. Second, a hybrid particle swarm optimization (PSO) integrated with trust region method (TRPSO) was proposed. The simulation results on some multimodal global optimizations showed that the algorithm is far more effective than linear decreasing weight PSO (LDWPSO) and sequential quadratic programming (SQP) to search the global optimum. Later, simultaneous strategies and the TRPSO were combined to solve the slurry high density polyethylene grade transition in a real plant. The detailed comparisons were derived from TRPSO, LDWPSO, iteration dynamic programming (IDP), and the real historical transition data of a plant demonstrated the effectiveness and practicability of the proposed method.