摘要
针对电子产品寿命预测中存在的不确定性因素影响,提出一种基于粒子滤波的电子产品动态损伤最优估计和寿命预测方法。首先建立了电子产品动态损伤HMM模型;分析了电子产品动态损伤和寿命预测中的不确定性因素;通过贝叶斯滤波模型,将寿命预测的不确定性问题转化为最优估计问题;利用粒子滤波算法求解出电子产品动态损伤的最优估计值,从而进行寿命预测;实验证明,该方法可有效消除系统和测量因素的干扰,明显提高电子产品剩余寿命预测的精度。
Aiming at the uncertainties in remaining life prediction of electronic products,a particle filter based method is proposed to assess dynamic damage and predict remaining life.A HMM model of dynamic damage is built firstly.The uncertainties in dynamic damage assessment and life prediction are analyzed.Using Bayesian filter model,the uncertainty problems in life prediction are transformed into optimal estimation problems,and particle filter algorithm is used to solve the optimal estimation value of electronic product dynamic damage.Experiment results prove that the proposed method can eliminate the uncertainties effectively and improve the life prediction precision distinctly.
出处
《仪器仪表学报》
EI
CAS
CSCD
北大核心
2011年第4期807-812,共6页
Chinese Journal of Scientific Instrument
基金
国防十一五重点预研项目(51317030103)资助
关键词
动态损伤
寿命预测
粒子滤波
不确定性
dynamic damage
life prediction
particle filter
uncertainty
