基于滤波器的一类带乘性噪声系统的执行器故障检测与估计方法

Actuator Fault Detection and Estimation Algorithm for a Class of System with Multiplicative Noise

本文针对一类新息可近似正态的带乘性噪声系统故障检测的必要性,给出了基于滤波器的一套执行器故障诊断算法。该算法首先对观测数据进行正态近似化判断,然后基于新息的近似正态分布特性,利用带乘性噪声系统的状态滤波算法判断故障的产生时刻,进而通过广义似然率法(GLR法)估计故障幅值。仿真验证了该算法的有效性。

In this paper,a set of filter-based actuator fault diagnosis algorithms were proposed based on the necessity of fault detection for a class of systems with multiplicative noise when the innovation is approximately normal.First,the observed data was judged by normal approximation.Then based on that,the time when faults appear was estimated by use of state filtering algorithms for systems with multiplicative noise.Finally the amplitude of faults was calculated by the generalized likelihood ratio（GLR） method.Simulation results show the validation of these algorithms.