摘要
支持向量机(SVM)一直被广泛应用于分类判别领域,在模拟电路的故障诊断中,电路普遍复杂多样,传统支持向量机只考虑数据类间距离最大化。本文中提出的粒子群优化的流形支持向量机,在保证数据最大类间间隔的同时,使映射在特征空间的数据,能保持原始空间的流形结构。同时将粒子群算法与SVM相结合,对支持向量机中的权重参数优化,使得对故障的诊断率比传统方法提高了2%～6%。通过实验发现,本文方法有效增强了模拟电路故障诊断的精确度。
The support vector machine(SVM)has been widely used for classification and discrimination,but the traditional SVM only considers the maximum distance between different data classes.To diagnose the fault of a complex and diverse analog circuit,this paper uses the manifold SVM based particle swarm optimization to ensure that the data mapped in the feature space can maintain the manifold structure of the original space while ensuring the maximum interval between different data classes.At the same time,the particle swarm optimization is used to optimize the weight parameters of the SVM,so that the fault diagnosis rate can be improved by 2%~6%compared with the traditional SVM.The experimental results show that this method effectively enhances the accuracy of analog circuit fault diagnosis.
作者
单剑锋
杨雨
Shan Jianfeng;Yang Yu(College of Electronic and Optical Engineering&College of Microelectronics,Nanjing University of Posts and Telecommunications,Nanjiing 210000,China)
出处
《机械科学与技术》
CSCD
北大核心
2019年第2期260-264,共5页
Mechanical Science and Technology for Aerospace Engineering
关键词
粒子群算法
支持向量机
流形结构
故障诊断
particle swarm optimization
support vector machine
manifold structure
analog circuit
fault diagnosis
