Informative Sensor Selection and Health Indicator Construction for Aircraft Engines Prognosis

Health condition of the engine directly affects the safety, reliability and efficiency of an aircraft. Prognostics enabling advanced alarming of failure and estimation of the remaining useful life has received increasing attention over the past decade. However, aircraft engines are precision systems with high uncertainty, it is difficult to model and predict the complex degradation of the engines. In this paper, a novel method for sensor selection and health indicator is proposed for the ef fective prognosis of the aircraft engine. The presented approach firstly selects informative sensors based on metrics of goodness, and then constructs synthesized health indicator by fusing the selected informative sensors. Case studies are implemented on the data set of an aircraft gas turbine engine. Results show that the proposed method can effectively fuse informative sensors to model the degradation of the aircraft engine.