摘要
根据1990—2011年中国历年石油消费量相关数据构造输入和输出向量,选用径向基函数(RBF)作为其函数,在MATLAB2.10工具箱中设置相应变量进行参数寻优,从而建立基于支持向量机的石油需求量预测模型.为了验证其效果,同时做出了最小二乘意义下的3种预测拟合曲线,数据误差分析结果表明,支持向量机模型的预测精度高、结果更为可靠.用支持向量机模型预测了2012—2015年我国的石油需求量.
A model for the prediction of China's oil demand is established using the regression algorithm of support vector machine.The radial basis function(RBF) is selected as the kernel function.China's oil consumption data from 1990 to 2006 are as training samples,and the data from 2007 to 2011 are as validation samples.The errors of the regression curve based on the SVM model and three regression curves based on least squares are analyzed.It is shown that the error of the regression curve based on the SVM model is the least.At last,China's oil demand from 2012 to 2015 is predicted using the trained support vector machine model.
出处
《西安石油大学学报(自然科学版)》
CAS
北大核心
2013年第2期103-106,2,共4页
Journal of Xi’an Shiyou University(Natural Science Edition)
基金
国家自然科学基金资助项目(编号:50875212)
高等学校博士学科点专项科研基金资助项目(编号:20126102130004)
教育部博士点基金资助项目(编号:200806990019)
关键词
石油需求量预测
支持向量机
回归算法
核函数
oil demand prediction
support vector machine
regression algorithm
kernel function