基于文化微粒群Elman网络的航煤干点软测量模型  

Soft-sensor Model of Jet Point Based on Cultural Particle Swarm Optimization Elman Network

作　　者：陈国初[1] 俞金寿[2] 

机构地区：[1]上海电机学院电气学院,上海200240 [2]华东理工大学自动化研究所,上海200237

出　　处：《系统仿真学报》2011年第10期2112-2117,共6页Journal of System Simulation

基　　金：上海市教委重点学科(J51901);上海市教委科研创新重点项目(09ZZ211);上海市闵行区科技项目(2010MH169)

摘　　要：将微粒群算法用于文化算法种群空间的优化,形成文化微粒群算法,并用常用测试函数检验该算法的性能;结果表明,文化微粒群算法具有比基本微粒群算法更好的优化性能。然后,将文化微粒群算法用于Elman网络连接权值和阈值的寻优,构成文化微粒群Elman网络,并将其应用于加氢裂化航煤干点软测量建模。结果表明,此模型精度高,应用前景广阔。Combining particle swarm optimization algorithm （PSO） with cultural algorithm （CA）, cultural particle swarm optimization algorithm （CPSO） was proposed. Both CPSO and PSO were used to resolve the optimization problems of five widely used test functions, and the results show that CPSO has better performance than PSO. Then CPSO was used to optimize Elman neural network＇s weights and thresholds. And cultural particle swarm Elman neural network （CPSOENN） was constructed. Next, CPSOENN was applied in soft-sensor modeling in Jet point of hydrocracking fractionator. Experiment results show that the model based on CPSOENN has high precision and good performance.

关 键 词：文化算法 微粒群算法 ELMAN网络 加氢裂化 航煤干点 软测量 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置] TQ206[自动化与计算机技术—控制科学与工程]