一种新型探测器在火灾探测中的应用  

The Application of A New Type of Detector in Fire Detecting

作　　者：戴丽[1] 张森强 

机构地区：[1]河南理工大学电气工程与自动化学院,河南焦作454000 [2]云南省易门供电有限公司,云南玉溪653100

出　　处：《电子质量》2008年第8期47-49,共3页Electronics Quality

摘　　要：提出将串联型模糊神经网络应用于火灾探测系统,并且使用多传感器技术。首先将多传感器测得的多个信号模糊化后送入神经网络进行处理,其模糊系统用指派法定义隶属度函数,神经网络采用3层前馈BP网络结构,用BP算法进行网络参数的训练。其后由模糊逻辑判决火灾或非火灾。The paper proposes the application of Series Fuzzy Neural Networks in fire detecting systems and the use of multi-sensor technology. First blur numbers of signals detected by multi-sensor and then they are sent into the neural network to carry out processing. Its fuzzy system applies the assignment method grade of membership function. Its neural network adopts three-tier feedforward BP network structure and practices the network parameters with BP algorithm. And finally judge fire or non-fire by fuzzy logic.

关 键 词：多传感器 火灾探测 模糊逻辑 神经网络 

分 类 号：TU892[建筑科学]