摘要
阐述了模糊支持向量机的原理。考虑到各类样本的不同影响,在基于风险的基础上,通过确定各个样本的模糊隶属度,引入模糊支持向量机的概念,将这一理论应用于汽轮机减速箱的故障诊断中,使一类样本点中只包含正常样本。实验结果表明了该方法的可行性与有效性。
The theory of Support Vector Machines is presented. Considering the different effects of input points, the conception of Fuzzy Support Vector Machines (FSVM) is applied in this article. The fuzzy membership to each input points is confirmed on the base of risk. The theory is applied for the fault diagnosis in the reducer of turbogenerator unit with only normal data involved. The experiment results proved the feasibility and validity of this method.
出处
《华北电力大学学报（自然科学版）》
CAS
北大核心
2003年第4期47-50,共4页
Journal of North China Electric Power University：Natural Science Edition
