关键词: 非色散红外, 偏最小二乘法, CO2气体浓度监测, 飞机火警探测器, 温度补偿模型, 浓度计算模型, 自适应报警算法

Abstract: In this paper, we analyze production rule of gaseous combustion products in fires, and put forward CO2 as parameter of fire detection. Through accurate monitoring of CO2 concentration in the environment, the aircraft fire detection has been realized. Considering the particularity of environmental condition on aircraft, NDIR（Non-Dispersion Infrared）is used to resolve the challenge of temperature compensation in aircraft fire detection. According to the relationship among the measured voltage of aircraft fire detector, environment temperature and reference voltage, temperature compensating model for aircraft fire detector is established based on partial least squares (PLS) approach. And model for CO2 concentration calculating is built by using partial least squares (PLS) method, which is based on correlation between measured voltage of aircraft fire detector and CO2 concentration. With the calculated concentration of CO2, an alarm algorithm for aircraft fire detector is presented then, which provides a new fire detection method for aircraft.

Key words: Non-Dispersion Infrared, partial least squares, CO2 concentration monitoring, aircraft fire detector, temperature compensating model, concentration calculating model, adaptive alarm algorithm