| 基于相空间重构和独立分量分析的火灾预测 |
| |
| 引用本文： | 刘盛鹏,张烨.基于相空间重构和独立分量分析的火灾预测[J].自然灾害学报,2011(3). |
| |
| 作者姓名： | 刘盛鹏 张烨 |
| |
| 作者单位： | 公安部上海消防研究所;南昌大学电子信息工程系; |
| |
| 基金项目： | 公安部应用创新计划项目(2009YYCXSHXF148) |
| |
| 摘 要： | 利用非线性动力学的相空间重构理论和独立分量分析,结合非线性支持向量回归,提出了火灾起数时间序列预测方法。首先用时间延迟法和独立分量分析重构系统的相空间来反映火灾起数时间序列的内在变化规律,然后再用非线性支持向量回归来构建预测模型。仿真结果表明该预测方法具有较高的预测精度。
|
| 关 键 词： | 火灾预测 独立分量分析 支持向量回归 相空间重构 |
Fire prediction based on phase space reconstruction and independent component analysis |
| |
| LIU Sheng-peng,ZHANG Ye.Fire prediction based on phase space reconstruction and independent component analysis[J].Journal of Natural Disasters,2011(3). |
| |
| Authors: | LIU Sheng-peng ZHANG Ye |
| |
| Institution: | LIU Sheng-peng1,ZHANG Ye2(1.Shanghai Fire Research Institute of Ministry of Public Security,Shanghai 200438,China,2.Department of Electronic and Information Engineering,Nanchang University,Nanchang 330031,China) |
| |
| Abstract: | Based on a combination of phase-space reconstruction of nonlinear dynamics with independent component analysis(ICA),a fire time series forecasting model was proposed by using support vector regression.The phase space of fire time series was reconstructed with the time delay and ICA,and then nonlinear support vector regression was used to construct the forecasting model.simulation results show that the proposed model can improve the fire time series forecasting accuracy. |
| |
| Keywords: | fire prediction independent component analysis(ICA) support vector regression(SVR) phase space reconstruction |
| 本文献已被 CNKI 等数据库收录！ |
