A method of health evaluation for the gear of helicopter main gearbox

Health evaluation of the helicopter main gearbox is essential for flight safety, so many condition indicators are got by processing the vibration data, but it is difficult to evaluate the health status accurately because of the variation of operation condition. In the paper the new condition indicators are obtained by the observer of multi-layer neural network based on the analysis of the effect of the flight parameters. A health evaluation model is established and the health indicator of the gear is brought forward through fusion of multi-indicator based on the Mahalanobis-Taguchi system. The health indicator proves to be efficient to evaluate the health status of the pinion of the main gearbox test-rig of helicopter.