高级检索

    王磊, 黄道. 基于改进的自组织映射网络的化工过程故障分类辨识[J]. 华东理工大学学报(自然科学版), 2006, (9): 1109-1112.
    引用本文: 王磊, 黄道. 基于改进的自组织映射网络的化工过程故障分类辨识[J]. 华东理工大学学报(自然科学版), 2006, (9): 1109-1112.
    WANG Lei, HUANG Dao. Fault Classification and Identification in Chemical Processes Based on Advanced SOM Algorithm[J]. Journal of East China University of Science and Technology, 2006, (9): 1109-1112.
    Citation: WANG Lei, HUANG Dao. Fault Classification and Identification in Chemical Processes Based on Advanced SOM Algorithm[J]. Journal of East China University of Science and Technology, 2006, (9): 1109-1112.

    基于改进的自组织映射网络的化工过程故障分类辨识

    Fault Classification and Identification in Chemical Processes Based on Advanced SOM Algorithm

    • 摘要: 将自组织映射网络(SOM)应用于化工过程故障数据的分类辨识,并采用粒子群优化(PSO)算法优化权重失真指数(LW D I),代替SOM的启发式训练算法,形成粒子群优化的SOM(PSO-SOM)分类算法。以某工厂甲醇合成反应器数据为研究对象,研究结果表明:对比基本SOM算法,PSO-SOM算法对复杂的故障数据能够得到较优的分类辨识结果,对甲醇合成生产中的故障诊断有非常显著的指导作用。

       

      Abstract: SOM algorithm is applied to fault data classification and identification in chemical processes.Direct optimization of 'locally weighted distortion index' by particle swarm optimizer(PSO) algorithm is substituted for Kohonen's heuristic-based training algorithm in SOM.A practical application of the new PSO-SOM algorithm in identifying and classifying fault data of methanol synthesis reactor is provided.The emulational experimental results show this algorithm can deal with more complicated data,obtain better classification results and identify fault's type more correctly than basic SOM algorithms.At the same time its realization is easier and simpler and is fit for scientific calculation and engineering application.Classification results can more greatly direct the optimization of methanol synthesis reactor parameters and yield monitoring.

       

    /

    返回文章
    返回