论文部分内容阅读
针对民用飞机光电感烟火灾探测器受探测机理及结构设计的限制,容易出现火警误报的问题,对该型探测器作出改进。用波长为670、980、1 060nm的三束激光作为光源,用消光系数比作为火灾探测机理设计感烟室,用RBF神经网络进行火灾信号智能识别,建立火灾烟雾识别模型。火灾模拟试验表明:新型光电感烟火灾探测器能够有效区分火灾烟雾颗粒和非火灾烟雾颗粒,是降低火警误报率的有效模型。
Aiming at the limitation of detection mechanism and structural design of photoelectric airborne smoke fire detectors of civil aircraft, the problem of fire false alarm is easy to occur and improvements are made to this type of detector. Three laser beams with wavelength of 670,980 and 1 060 nm were used as light source, and the smoke chamber was designed with extinction coefficient ratio as the fire detection mechanism. The fire signal was identified by RBF neural network intelligently, and a smoke smoke recognition model was established. Fire simulation tests show that the new photoelectric smoke detector can effectively distinguish between fire smoke particles and non-fire smoke particles, which is an effective model to reduce the false alarm rate of fire alarm.