Test point optimization for model-based fault diagnosis of satellite aviation system

Abstract

To realize autonomous control of satellite, NASA uses the method of autonomous fault diagnosis. In the fault diagnosis, it is found that the fault diagnosis accuracy and the fault coverage have relation with the choice of test points. In order to solve the optimization problem of test points in the fault diagnosis of satellite aviation system, this paper presents an optimization method of test points. This method constructs correlation matrix firstly based on functional block diagram and signal flow chart of system, then uses the method of decomposing matrix gradually through analyzing fault detection and fault isolation to select the optimal test points. And after considering reliability and cost, it can achieve the purpose of improving fault accuracy and reducing test costs. The fault diagnosis results for a typical satellite aviation components show that the proposed method is effective.