扩散式遗传-微粒群优化算法及其应用  

ScGA-PSO Algorithm and Its Application

作　　者：林晨[1] 俞金寿[1] 

机构地区：[1]华东理工大学自动化研究所,上海200237

出　　处：《石油化工自动化》2009年第1期23-26,共4页Automation in Petro-chemical Industry

摘　　要：将PSO算法与ScGA(扩散式遗传算法)相结合,提出ScGA-PSO优化算法。通过对4种常用的测试函数进行优化和比较,结果表明ScGA-PSO性能相比PSO有明显的提升,且更容易找到最优解。然后将ScGA-PSO用于延迟焦化装置主分馏塔汽油干点软测量,建立基于ScGA-PSO的粗汽油干点神经网络软测量模型,通过与实际工业数据对比,表明该模型精度高、性能好,具有广阔的应用前景。An improved particle swarm optimization algorithm -- ScGA-PSO was proposed based on PSO and Scatter GA. Then ScGA-PSO and PSO were used to resolve four widely used test functions＇ optimization problems. Results show that ScGA-PSO＇s performance is much better than PSO＇s and ScGA-PSO can find the best fit easier. Next ScGA-PSO is applied to train artificial neural network to construct a practical soft-sensor of gasoline endpoint of main fractionator of delayed eoking unit. The obtained results and comparison with actual industrial data indicate that the new method proposed by this paper is feasible and effective in soft-sensor of gasoline endpoint.

关 键 词：微粒群优化算法 扩散式遗传算法 粗汽油干点值 软测量 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]