Analysis of Condition-based Calibration for TDLAS Oxygen Sensors in Aircraft Fuel Tanks

The flammability of aircraft fuel tanks is a significant safety concern. Oxygen sensors based on Tunable Diode Laser Absorption Spectroscopy (TDLAS) are used for measuring oxygen concentration in fuel tanks, ensuring that the inerting system maintains safe oxygen levels. However, over time, TDLAS oxygen sensors experience laser emission intensity decay and gas cell contamination, which degrade measurement accuracy. There has been little research on this topic. In response, this paper proposes that the degradation of the laser and the contamination of the gas cell can be monitored by observing the laser emission intensity and the ratio of the transmitted laser intensity through the gas cell to the emission light intensity in the non-absorption region. It also suggests using theoretical formulas or a GA-BP neural network to correct errors in oxygen concentration measurements within a specified threshold range, with maintenance or replacement only occurring when the threshold is exceeded. This approach ensures measurement accuracy while avoiding unnecessary downtime or maintenance. Simulation data demonstrate the effectiveness of the GA-BP neural network in calibrating oxygen concentration measurement errors. The research findings provide a new approach for the effective utilization of TDLAS oxygen sensors in domestically-produced aircraft.