基于噪声滤波的非线性系统故障诊断方法研究

Research on System Fault Diagnosis Method Based on Correlation Noise Filtering[JZ)][HS)]

摘要:

随着复杂系统故障诊断的要求不断增加,非线性滤波技术复杂系统诊断已越来越成为研究的热点与难点问题之一；针对传统的粒子滤波进行系统突变故障诊断的问题,文章提出了一种改进的噪声粒子滤波故障诊断新方法；方法给出了噪声粒子滤波统计模型,通过粒子滤波得到状态估计值,并得到全概率分布信息用于故障检测中；最后通过仿真实验以及数值模拟验证了文章提出的方法在复杂系统故障诊断中是有效的,同时具有较高的精确性。

Abstract:

with the fault diagnosis of complex system requirements increase, nonlinear filtering technique for complex system diagnosis has increasingly become one of the hot and difficult problems in research.Aiming at the systematic mutation fault diagnosis problem in traditional particle filter is presented, this paper presents a new method of noise improved particle filter fault diagnosis. The noise statistical model method of particle filter, the state estimation is obtained by the particle filter, and is used for fault detection of information distribution in all probability. Finally, through the simulation experiment and numerical simulation demonstrate that the proposed methods are effective in the fault diagnosis of complex system, and have high accuracy.