对于含有奇异点的系统而言,由于一般的剪枝算法不能成功抑制系统中的奇异点,在借鉴支持向量分类机选择支持向量方法的基础上,提出了改进的剪枝算法.改进的剪枝算法能成功抑制系统中的奇异点,减少支持向量的数目,增强最小二乘支持向量回归机的泛化能力.另外,仿真实例也验证了改进剪枝算法的有效性.在不含有奇异点系统中,改进的剪枝算法退化成了一般的剪枝算法,也就是说一般的剪枝算法是改进剪枝算法的一个特例. For systems with singular points, an improved pruning algorithm is proposed based on the support vector machine (SVM) based on SVM because the general pruning algorithm can not suppress the singularity in the system successfully. The algorithm can successfully suppress the singularities in the system, reduce the number of support vectors and enhance the generalization ability of LSSVM. In addition, the simulation example also verifies the effectiveness of the improved pruning algorithm.In the system without singularities , The improved pruning algorithm degenerates into a general pruning algorithm, that is to say, a general pruning algorithm is a special case of improving the pruning algorithm.