中文摘要：

在两个方面对支持向量机进行改进：针对最小二乘支持向量机缺失稳健性问题,建立稳健LS-SVM模型,通过仿真试验验证该模型的正确性和可靠性;结合支持向量机的数学性质,提出新的动态LS-SVM算法,最后将两者结合形成动态稳健LS-SVM模型,并用此模型进行大坝变形预测,取得了较好的预测效果。

英文摘要：

In view of the deficiencies of support vector machine （SVM） , it has been improved in two aspects. On the one hand, the paper presents a robust LS-SVM model and its accuracy and reliability are verified through the simulation test. On the other hand, with respects of the nature of mathematics of SVM, a new dynamic LS-SVM method is proposed. Finally, the dynamic robust LS-SVM model is formed based on the combination of the dynamic LS-SVM and the robust LS-SVM, and applied to the prediction of dam deformation, which has achieved good prediction results.