中文摘要：

提出一种新的属性约简算法——基于符号有向图的属性约简算法。该算法利用符号有向图中故障只能在相容通路中传播的性质对系统中的变量进行约简,然后利用支持向量机优良的分类性能进行故障诊断。挑选过程控制界公认的Tennessee-Eastman Process仿真系统进行属性约简,并进行故障诊断仿真,证明了该算法的可行性,该方法可广泛应用于化工和冶金等大型系统的故障诊断中。

英文摘要：

A new attribute reduction method,SDG-based attribute reduction method,is proposed in this paper.This method reduces the variables of fault diagnosis system by the characteristic of the consistent path that is the fault only spread in the consistent path.SVM is used to diagnose fault because of its good classification performance.The Matlab simulation of Tennessee-Eastman Process（TEP） demonstrates the feasibility of the attribute reduction method.This can be widely applied to fault diagnosis of the chemical and metallurgical industry.