摘要
针对煤炭需求量数据的小样本、非线性特点,提出了一种基于超球向量机的煤炭需求量预测方法。首先采用类别分析法选择煤炭需求量的影响因子,然后通过超球向量机对预测的参数进行优化,最后建立煤炭需求量与隶属度因子之间复杂的非线性关系模型。结果表明,相对于参比模型,改进超球向量机提高了煤炭需求量的预测精度,能够准确刻画煤炭需求量变化趋势。
In view of coal requirement's small sample data,nonlinear characteristic,this paper proposes coal requirement prediction method based on improved least squares support vector machines.Firstly,the influence factors of coal requirement area are selected by multiple regression analysis method,and then the parameters of least square support vector machines are optimized by genetic algorithm,lastly build the complex nonlinear model between coal requirement and influence factors.The results show that the proposed model has improved the prediction accuracy of coal requirement compared with other prediction models;the proposed model is an effective forecasting method for coal requirement
出处
《科技通报》
北大核心
2013年第6期25-26,29,共3页
Bulletin of Science and Technology
基金
河南省教育厅科学技术研究重点项目(12B520075)
关键词
煤炭需求
影响因子
超球向量机
coal requirement
prediction model
least square support vector machines
genetic algorithm