基于核主元分析与多支持向量机的监控诊断方法及其应用

Monitoring model based on kernel principal component analysis and multiple support vector machines and its application

为保证密闭鼓风炉冶炼过程的正常运行,构造了一种基于核主元分析(KPCA)和多支持向量机(MSVM)的监控模型.该监控模型首先用核主元分析方法对过程数据进行特征提取,然后将代表过程特征的核主元送入到多支持向量机分类器中进行故障诊断与分类.仿真研究显示,该监控模型具有较好的泛化能力,能有效地应用于鼓风炉的监控诊断,可用于鼓风炉熔炼过程的现场操作指导.

In order to keep the imperial smelting furnace （ISF） working smoothly, a dynamical monitoring model based on kernel principal component analysis （KPCA） and multiple support vector machine （MSVM） is constructed. Firstly, the KPCA approach is adopted to extract the feature, and then the SVM multiclass classifiers are used for classification with the input of the feature. The simulation research shows that this dynamical monitoring model has the performances like, better generalization, stronger robust, and preferable monitoring result. Now, the monitoring model is applied in the assistant decision-making system of a imperial smelting furnace.