基于粒子滤波优化的滚动式时间序列多步预测  

Multi-step prediction of rolling time series based on particle filter optimization

作　　者：杨淑莹[1] 王丽贤[1] 牛廷伟[1] 邓飞[2] 

机构地区：[1]天津理工大学智能计算及软件新技术重点实验室,天津300384 [2]佛罗里达国际大学工程和计算机学院,美国迈阿密33174

出　　处：《系统工程与电子技术》2012年第6期1097-1101,共5页Systems Engineering and Electronics

基　　金：国家自然科学基金(61001174)资助课题

摘　　要：针对复杂的应用环境下,时间序列建模不易准确,多步预测精度不高的问题,提出基于粒子滤波(particle filter,PF)优化的滚动式时间序列(roll time series,RTS)多步预测算法(PF_RTS)。采用Box-Jenkins方法对时间序列滚动自适应建模,所建模型作为粒子的状态转移方程,利用粒子滤波算法实时动态修正预测数据,逼近状态的最优估计。本文算法具有自学习能力,适合实时应用。仿真结果表明,本文算法需要的先验知识少,提高了预测的精度。For complex application environments, it is difficult to get accurate time series modeling and multi-step prediction results. Multi-step prediction of rolling time series based on the particle filter optimization（PF_RTS） is proposed to solve the problem. According to the modeling thoughts of Box-Jenkins, the time se- ries is adaptively modeled. And the model is regarded as the particles~ state transition equation. By using a par-ticle filtering algorithm, the optimal state is estimated and the predicted data are real-time corrected. With selflearning ability, this algorithm is suitable for real-time applications. The simulation results show that thismethod needs less prior knowledge and has a better predictive accuracy.

关 键 词：时间序列 多步预测 粒子滤波 Box-Jenkins建模 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]