摘要
支持向量机学习算法针对小样本情况表现出优良的性能,能够在有限特征信息情况下,最大限度地发掘数据中隐含的分类知识,使其能够更适用于故障诊断领域。研究决策有向无环图多类分类支持向量机在TE(Tennessee Eastman,TE)过程中的应用。仿真结果表明该方法分类精度较高且测试时间短,能够满足复杂工业过程对故障诊断的要求。
The support vector machines shows fine performance to limited samples and it can acquire connotative class information to great extent from limited samples which made it more favorable in fault diagnosis field. The support vector machine with decision directed acyclic graph is used in tennessee eastman process. The experimental results show that its classification accuracy rate is high, and test time is short. It can meet the demands of complicated industrial process.
出处
《安徽工业大学学报（自然科学版）》
CAS
2010年第2期195-199,共5页
Journal of Anhui University of Technology（Natural Science）
关键词
支持向量机
故障诊断
TE过程
support vector machine
fault diagnosis
tennessee eastman process
