论文部分内容阅读
提出一种基于序贯二次规划(SQP)法的混沌粒子群优化方法(CPSO-SQP).将混沌PSO作为全局搜索器,并用SQP加速局部搜索,使得粒子能够在快速局部寻优的基础上对整个空间进行搜索,既保证了算法的收敛性,又大大增加了获得全局最优的几率.仿真结果表明,算法精度高、成功率大、全局收敛速度快,明显优于现有算法.将所提出的算法用于高密度聚乙烯(HDPE)装置串级反应过程的乙烯单耗优化,根据工业反应机理以及现场操作经验分析可知,所提出的算法是可行的.
A method based on Sequential Quadratic Programming (SQP) for chaos particle swarm optimization (CPSO-SQP) is proposed. Chaos PSO is used as a global searcher and SQP is used to speed up local search so that particles can be quickly and locally optimized The search of the whole space not only guarantees the convergence of the algorithm but also greatly increases the probability of getting the global optimum.The simulation results show that the algorithm has the advantages of high precision, high success rate and fast global convergence, which is obviously superior to the existing algorithms The proposed algorithm is used to optimize the ethylene consumption in cascade reaction of HDPE. According to the industrial reaction mechanism and field operation experience, the proposed algorithm is feasible.