基于随机摄动粒子滤波器的故障预报算法  

Fault prediction algorithm based on stochastic perturbation particle filter

作　　者：张琪[1] 胡昌华[1] 乔玉坤[1] 蔡艳宁[1] 

机构地区：[1]第二炮兵工程学院302教研室,西安710025

出　　处：《控制与决策》2009年第2期284-288,共5页Control and Decision

基　　金：国家自然科学基金重点项目(60736026);教育部新世纪优秀人才支持计划项目(NCET-07-0144)

摘　　要：针对粒子滤波算法在故障预报中的大计算量和粒子退化问题,提出一种基于随机摄动粒子滤波器的故障预报算法.当粒子退化严重时,对粒子用随机摄动方式进行再采样,一方面可改进样本的多样性,缓解粒子退化;另一方面可缩短再采样时间,减少计算量,从而提高粒子滤波算法的跟踪能力.仿真结果表明该算法可行,能及时准确地对系统故障进行预报.Degeneracy of particles and large computing cost are the main problems when particle filters are applied to fault predictions. Therefore, a fault prediction algorithm based on stochastic perturbation particle filter is proposed to resolve the above problems. The stochastic perturbation re-sampling is used when the degeneracy of particles is serious, which can improve the diversity of samples, ameliorate the degeneracy of particles, shorten the re-sampling time and reduce the computing cost. As a result, the.tracking ability of particle filter is improved. Simulation results demonstrate that the algorithm proposed is valid and the system fault can be predicted accurately and timely.

关 键 词：粒子滤波 退化现象 计算量 随机摄动 故障预报 

分 类 号：TP13[自动化与计算机技术—控制理论与控制工程]