The invention discloses a TCAS
system fault comprehensive diagnosis method based on a Bayesian
decision tree. The TCAS
system fault comprehensive diagnosis method comprises the following steps that firstly, the fault
decision tree is used for decomposing
airplane electronic
system faults, and a measuring method is provided; secondly, according to requirements of a measured or controlled object, parameter information such as a
voltage, a current, frequency, a
harmonic wave and resistance are collected, and analog-to-
digital conversion is carried out; thirdly,
feature extraction is carried out on the
voltage, the current and the frequency, and state information of devices is obtained according to measurement of extracted power source quality features or the
voltage or the resistance; fourthly, the faults of each device are judged with a statistics-based Bayesian decision method; fifthly, the object-oriented
decision tree is applied to
inference of fault locations, the faults are removed one by one, and fault positioning is carried out; sixthly, a detection result is output. Possible hidden dangers of an
airplane can be detected while the faults are removed, and a TCAS electronic system fault diagnosis system with an industrial control computer serving as a host computer is achieved.