| 峰值识别的SVM模型及在时用水量预测中的应用 |
| |
| 引用本文： | 俞亭超,张土乔,柳景青.峰值识别的SVM模型及在时用水量预测中的应用[J].系统工程理论与实践,2005,25(1):134-139. |
| |
| 作者姓名： | 俞亭超 张土乔 柳景青 |
| |
| 作者单位： | 浙江大学土木工程学系 |
| |
| 基金项目： | 国家自然科学基金(50078048) |
| |
| 摘 要： | 提出了一种峰值识别理论及相应改进的SVM模型.该模型在结构风险最小化准则的目标函数中加大峰值误差的权重,并结合杭州市时用水量预测实践结果,其能很大提高对峰值用水量的预测精度.模型还把时用水量序列分为参数优化集和训练预测集,以对参数C、γ优化选取.
|
| 关 键 词： | 支持向量机（SVM） 时用水量 预测 |
| 文章编号： | 1000-6788(2005)01-0134-07 |
| 修稿时间： | 2003年11月17 |
SVM Model with Peak Value Recognition and Its Application to Hourly Water Consumption Forecasting |
| |
| YU Ting-chao,ZHANG Tu-qiao,LIU Jing-qing.SVM Model with Peak Value Recognition and Its Application to Hourly Water Consumption Forecasting[J].Systems Engineering —Theory & Practice,2005,25(1):134-139. |
| |
| Authors: | YU Ting-chao ZHANG Tu-qiao LIU Jing-qing |
| |
| Institution: | Department of Civil Engineering, Zhejiang Universi |
| |
| Abstract: | Water consumption forecasting is the basis and precondition for the optimization and operation of water supply system, whose accuracy influences directly the reliability and feasibility of control and operation. A modified support vector machine (SVM) model with peak recognition theory is proposed. By this model weight to the error of big value was input to loss function of structural risk minimization. According to the hourly water demand forecasting results of Hangzhou city, the accuracy of peak value for... |
| |
| Keywords: | support vector machine hourly water consumption forecasting |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《系统工程理论与实践》浏览原始摘要信息 |
| 点击此处可从《系统工程理论与实践》下载免费的PDF全文 |
