A separation hyperplane is constructed based on support vector domain description(SVDD),which attempts the combination of SVDD with SVM.
针对两类分类问题中使用支持向量机(Support Vector Machines,SVM)训练时间长和支持向量域分类器(Support Vector Domain Classifier,SVDC)精度不高的问题,建立一种基于支持向量域描述(Support Vector Domain Description,SVDD)的分离超平面,尝试将SVDD与SVM结合。
The data of every fault class are trained by support vector domain description (SVDD)to obtain the optimal enclosing feature spaces.
该模型采用支持向量域描述算法(SVDD)对多类故障进行单独训练,建立独立而封闭的特征空间,满足故障类别的动态增加需要,并采用样本与各特征空间的相对距离进行了多故障的混合识别。