改进PSO算法在氧化动力学模型参数估计中的应用  

Application of Modified Particle Swarm Optimization Algorithm in Parameter Estimation for Oxidation Kinetics Model

作　　者：陈富 伍铁斌[2,3] 殷永生 成运[2] 刘云连[2] 

机构地区：[1]江西应用工程职业学院电气工程系,萍乡337042 [2]湖南人文科技学院通信与控制工程系,娄底417000 [3]中南大学信息科学与工程学院,长沙410083

出　　处：《科学技术与工程》2013年第16期4481-4485,4490,共6页Science Technology and Engineering

基　　金：国家自然科学基金项目(61273159);湖南省自然科学基金项目(12JJ2040);湖南省重点建设学科项目;湖南人文科技学院青年基金项目(2012QN07)资助

摘　　要：通过构造一个适当的目标函数,将Hg氧化动力学模型的参数估计问题转化为一个多维数值优化问题;然后提出一种基于算术交叉和多样性变异的改进PSO算法来求解该优化问题。算法随机选择粒子与当前最优粒子进行算术交叉操作,将粒子逐步向极值点引导,提高算法的局部搜索能力。引入多样性变异算子以维持种群粒子的多样性。几个标准测试函数的实验结果表明算法具有较好的寻优性能。将算法应用于Hg氧化动力学模型参数估计中,获得了满意的结果。Through establishing an appropriate objective function, the parameter estimation problem for Hg oxi- dation kinetic model was formulated as a multi-dimensional numerical optimization problem, which can be solved by modified particle swarm optimization （CMPSO） algorithm. In the evolution process, arithmetic crossover operator is utilized for the randomly selected particle and the optimal particle. The crossover operator can lead gradually the population to the extreme point and improve the local searching ability. In addition, diversity mutation strategy is introduced to maintain the diversity of the particle. Several benchmark functions are tested ; the experimental results show that CMPSO has better optimization performance. CMPSO is applied to estimate the kinetic parameters of Hg oxidation, and a satisfactory result is obtained.

关 键 词：粒子群优化算法 算术交叉 多样性变异 动力学模型 参数估计 

分 类 号：O643.1[理学—物理化学] TP18[理学—化学]