基于粗糙集与支持向量机的故障智能分类方法

结合粗糙集的属性约简与支持向量机的分类功能，提出一种应用粗糙集与支持向量机的故障分类方法。该方法应用粗糙集理论属性约简作为诊断数据预处理器，可将冗余属性从诊断决策表中删除，而不损失有效信息，然后基于支持向量机进行故障分类建模和预测。谊方法可降低故障诊断数据维数及支持向量机在故障分类过程中的复杂度，但不会降低分类性能。将方法应用于某柴油机故障诊断数据的测试分类，结果表明该方法可快速正确的从数据获得故障类剐。

Method of Fault Intelligent Classification Based on Rough Set and Support Vector Machine

A method based on rough set and support vector machine and applied for fault classification is proposed in this paper. Using the rough set reduction algorithm as the pretreatment of diagnosis data, it can get rid of redundant attributes of decision table. Then support vector machine is used the to fault classification modeling and forecast after rough set reduction. The method can reduce the dimensions of the fault diagnosis data and the complexity of the fault classification with SVM, and can not affect its classification performance. Applied to fault data classification of diesel engine, it can obtain the fault class fast and actually.