摘要
在提取发动机气门机构故障特征的基础上,提出了采用粗糙集和支持向量机相结合的故障诊断方法。首先,基于粗糙集理论对故障诊断决策表进行属性约简,然后在最优决策属性的基础上使用支持向量机分类器对故障进行分类。实际诊断结果验证了采用粗糙集与支持向量机相结合的方法对故障进行诊断的可行性与有效性。
Based on fault feature extraction from engine valve-train, a new hybrid scheme of rough sets and support vector machine for fault diagnosis was proposed. Firstly, reduction was done based on rough sets theory, and the optimal diagnostic decision table was determined. Secondly, based on the optimal decision attributes, support vector machine classifier was used for fault classification. The results demonstrate that this method is effieient and feasible to fault diagnosis.
出处
《内燃机学报》
EI
CAS
CSCD
北大核心
2006年第4期379-383,共5页
Transactions of Csice
关键词
粗糙集
支持向量机
故障诊断
发动机
Rough sets
Support vector machine
Fault diagnosis
Engine
