对于小规模的非静态数据,最近提出的时间自适应支持向量机(TA-SVM)方法表现出良好的性能,它从兼顾局部优化和全局优化的角度同时求解多个子分类器的特性.但对于大数据集,较高的计算代价限制了它的实用性.针对此不足,结合核心向量机(CVM)理论提出了针对非静态大数据集的新颖分类方法,即基于中心约束最小包含球(CCMEB)的TA-CVM,简称CCTA-CVM.该方法具有渐近线性时间复杂度的优点,同时继承了TA-SVM的良好性能.最后通过实验验证了所提出方法的有效性. For small-scale, non-stationary data, the recently proposed time-adaptive support vector machine (TA-SVM) method shows good performance, which solves the characteristics of multiple sub-classifiers simultaneously from the perspective of both local optimization and global optimization. In order to solve this problem, a new classification method based on kernel vector machine (CVM) theory is proposed for non-static large data sets, that is, based on the Central Constraint Minimal Inclusion Ball (CCMEB ) TA-CVM, referred to as CCTA-CVM for short, has the advantages of asymptotic linear time complexity and inherits the good performance of TA-SVM.Finally, the effectiveness of the proposed method is verified by experiments.