| 演化追踪法优化相空间的SVM供水量预测模型 |
| |
| 引用本文： | 赵雪凝,杜坤,周明,任刚红,李诚.演化追踪法优化相空间的SVM供水量预测模型[J].土木建筑与环境工程,2016,38(Z2):147-150. |
| |
| 作者姓名： | 赵雪凝 杜坤 周明 任刚红 李诚 |
| |
| 作者单位： | 昆明理工大学 建筑工程学院，昆明 650500,昆明理工大学 建筑工程学院，昆明 650501,昆明理工大学 建筑工程学院，昆明 650502,昆明理工大学 建筑工程学院，昆明 650503,昆明理工大学 建筑工程学院，昆明 650504 |
| |
| 基金项目： | 昆明理工大学2016年学生课外学术科技创新基金(2015YB025)；国家自然科学基金（51608242）；云南省人才培养计划项目 (14118943) |
| |
| 摘 要： | 相空间重构的支持向量机预测模型应用十分广泛,在城市供水量预测方面也占据着重要地位,传统的预测模型趋向于将重构的相空间整体带入,这样可能存在引入无效相点从而影响预测精度的问题,基于此将演化追踪法引入相空间重构的预测模型对有效相点进行筛选,优化预测模型的训练样本,达到提高预测精度目的。利用MATLAB编程软件将演化追踪法用于城市供水量的预测,预测结果的平均绝对误差由0.52%降低到了0.29%,证明了演化追踪法的可利用性与有效性。
|
| 关 键 词： | 重构相空间 SVM 演化追踪法 供水量预测 |
| 收稿时间： | 2016/10/29 0:00:00 |
Improvement of SVM regression forecast water supply model based on phase space reconstruction |
| |
| Zhao Xuening,Du Kun,Zhou Ming,Ren Ganghong and Li Chen.Improvement of SVM regression forecast water supply model based on phase space reconstruction[J].土木建筑与环境工程,2016,38(Z2):147-150. |
| |
| Authors: | Zhao Xuening Du Kun Zhou Ming Ren Ganghong and Li Chen |
| |
| Affiliation: | Faculty of Civil Engineering and Mechanics, Kunming University of Science and Technology, Kunming 650500, P. R.China,Faculty of Civil Engineering and Mechanics, Kunming University of Science and Technology, Kunming 650501, P. R.China,Faculty of Civil Engineering and Mechanics, Kunming University of Science and Technology, Kunming 650502, P. R.China,Faculty of Civil Engineering and Mechanics, Kunming University of Science and Technology, Kunming 650503, P. R.China and Faculty of Civil Engineering and Mechanics, Kunming University of Science and Technology, Kunming 650504, P. R.China |
| |
| Abstract: | Nowadays, SVM prediction model based on phase space reconstruction is widely used, and it also plays an important role in urban water supply prediction. The traditional prediction model tends to bring the reconstructed phase space into the whole, which may lead to ineffective introduction of SVM. Phase prediction method is used to improve the accuracy of prediction. Based on this, the evolutionary tracing method is introduced into the prediction model of phase space reconstruction to filter the effective points and to optimize the training samples of the prediction model. The evolutionary tracing method is used to forecast the urban water supply quantity by using MATLAB programming software. The average absolute error of forecasting result is reduced from 0.52% to 0.29%, which proves the availability and effectiveness of evolutionary tracing method. |
| |
| Keywords: | reconstructed phase space SVM evolutionary tracing method water supply prediction |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《土木建筑与环境工程》浏览原始摘要信息 |
|
点击此处可从《土木建筑与环境工程》下载全文 |
|
