摘要

The engine health monitoring system has been recently applied to most advanced aircrafts to improve reliability and durability of the propulsion systems as well as to minimize the operational cost. Among aero engines, the helicopter engine is generally operated in much severe environmental conditions such as hot and cold temperature, snow, heavy rain, foreign particles such as sand and dust due to low altitude flight mode and rotor blade rotation. Therefore these operating conditions give rise to damages of engine gas path components.
Recently, health monitoring of major gas path components using the GPA (Gas Path Analysis) method has been performed in identifying and quantifying engine faults. Moreover in order to improve the GPA method, artificial intelligence methods such as Genetic Algorithms, Fuzzy Logic and Neural Network have been applied to gas turbine engine fault diagnostics.
In this study, an on-line diagnostic program of a helicopter turboshaft engine is proposed with an on-line condition monitoring program using SIMULINK and a fault diagnostic program using Neural Network. The on-line condition program can monitor the difference between the real measuring performance data and the base performance data calculated by the base engine performance model at measuring atmospheric and flight conditions. The fault diagnostic algorithm using Neural Network can identify the fault type of the gas path components and quantify the fault size through changes of component performance parameters from variation of measuring performance data using the on-line condition program.
Evaluation of on-line diagnostic program developed in this study is carried out through condition monitoring test using a micro gas turbine test system and detecting implanted faults of the helicopter turboshaft engine.

  • 出版日期2010