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。
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.
出处
《大地测量与地球动力学》
CSCD
北大核心
2009年第2期118-120,130,共4页
Journal of Geodesy and Geodynamics
基金
国家自然科学基金(40674008)
关键词
最小二乘支持向量机
稳健估计
动态预测
仿真实验
大坝变形
Least Square Support Vector Machine
robust estimation
dynamic prediction
simulation test
dams＇ deformation
