摘要
针对工业过程中某些重要过程变量难以实现实时在线检测和高维数据处理的问题,提出了将主元分析与最小二乘支持向量机相结合的软测量建模方法,并利用该方法建立了工业阿维菌素发酵过程中的菌丝浓度软测量模型.主元分析方法的引入,有效地提高了最小二乘支持向量机软测量模型的精度和泛化能力.应用结果表明,该方法与基于径向基函数神经网络软测量模型相比具有有效性和优越性.
To solve the problems of real-time on-line measurements of some important process variables and of data handling with high dimension, a novel method of soft sensor based on the integration of both PCA and LS-SVM is proposed and applied in the modelling of biomass estimation in the fermentation process of Avermectin in this paper. The introduction of the method of PCA contributes to the distinct improvement of precision and generalization ability of the soft sensor model based on LS-SVM. Industrial application has shown that the proposed method is superior to the soft sensor model based on RBF neural network.
出处
《江南大学学报（自然科学版）》
CAS
2006年第2期182-186,共5页
Joural of Jiangnan University (Natural Science Edition)
基金
国家自然科学基金项目(20206027)
国家973计划项目(2002CB312200)
关键词
最小二乘支持向量机
主元分析
软测量
RBF神经网络
least squares support vector machine （LS-SVM）
principal component analysis （PCA）
soft sensor
RBF neural network
