基于RBF网络的火灾探测信号处理算法研究  

Signal Processing Algorithm for Fire Detection Based upon RBF Neural Network

作　　者：张春生[1] 赵云胜[1] 卢颖[1] 叶彬[1] 

机构地区：[1]中国地质大学工程学院,武汉430074

出　　处：《安全与环境工程》2010年第6期51-54,共4页Safety and Environmental Engineering

摘　　要：本文通过建立火灾探测信号处理的RBF网络模型,并与通常的BP网络模型和期望结果进行对比,结果表明:在进行火灾探测信号处理中,RBF网络可避免BP网络的局部极小以及收敛速度慢等缺点,在精度、训练速度等方面均优于BP网络。该研究为处理火灾探测信号等非结构问题提供了一种行之有效的方法。A radial basis function（RBF） neural network model for the data processing technique of the recognition of fire smoke signal is established.The signal processing model based on radial basis function network is studied through experimental data and verified by additional data successfully.This paper also compares another network model based on back propagation network with the RBF model.The results show that the radial basis function network is much better than back propagation network in accuracy and training speed for the problem studied.By the simulation experiments,the method of RBF neural network is proved to be an effective way to deal with fire smoke detecting or non-structural issues.

关 键 词：火灾探测 信号处理 算法 径向基函数 人工神经网络 

分 类 号：X928.7[环境科学与工程—安全科学]