摘要
提出了支持向量机和证据理论结合的思想;建立了支持向量机初步诊断与证据理论融合决策诊断相结合的信息融合故障诊断模型;给出了基于SVM的基本概率分配构造方法和诊断决策规则。以200 MW汽轮发电机组故障诊断为例,充分地验证了该方法的诊断性能。
The method of combining Support Vector Machine （SVM） with D-S evidential theory is presented in the paper. A fault diagnosis model for information fusion is constructed, which combines the SVM based preparatory diagnosis with the decision diagnosis based on D-S evidential theory. The basic probability assignment method and the rules of diagnosis decision-making based on SVM are given. Taking the fault diagnosis of turbine generator set for example, the performance of the new method is validated.
出处
《电光与控制》
北大核心
2007年第4期187-190,共4页
Electronics Optics & Control
关键词
支持向量机(SVM)
证据理论
信息融合
故障诊断
Support Vector Machine （SVM）
D-S evidential theory
information fusion
fault diagnosis
