摘要
提出了一种基于支持向量机回归算法的港口吞吐量非线性组合建模预测方法,并运用该方法进行了港口吞吐量预测,同时将该预测结果与其他方法的预测结果进行了比较.结果表明,该方法具有很强的学习及泛化能力,在处理具有一定程度的不确定性的非线性系统的组合建模预测问题时具有很好的应用价值.
A nonlinear combined model for prediction of port throughput was presented based on the regression algorithm of support vector machines. The comparison of the predicted result of port throughput by the present method with those of other methods shows that the method developed in this paper is of strong capability in learning and mapping, and has high applicability in modeling for combined prediction of nonlinear systems with some uncertainties.
出处
《河海大学学报(自然科学版)》
CAS
CSCD
北大核心
2005年第4期482-484,共3页
Journal of Hohai University(Natural Sciences)
基金
上海市高等学校科学技术发展基金资助项目(02IK14)
关键词
港口吞吐量
非线性组合预测
支持向量机回归
<Keyword>port throughput
nonlinear combined prediction
support vector machine regression