一种基于BP神经网络的建筑防火检测改进方法

An Improvment Methods for Building Fire Detection Based on BP Neural Network

针对现有的建筑防火检测方法已经无法满足在实际检测中需求的问题,本文提出了一种基于建筑防火检测的改进BP神经网络模型。在传统BP神经网络的基础上,提出动态合并与删减规则,并且根据建筑防火检测的需求建立检测的指标,再根据改进BP神经网络和检测指标建立检测模型。仿真实验表明,基于建筑防火检测的改进BP神经网络模型实际操作性很强,可以应用于对建筑物的防火检测,值得推广使用。

According to the problem of the existing building fire detection methods can not meet the actual needs of the detection, this paper proposes an improved BP neural network model based on building fire detection. And on the basis of the traditional BP neural network, the paper proposes the dynamic combination and exclusion rules, establishes detection index according to the building fire detection needs, and finally establishes detection model based on improved BP neural network and testing index. Simulation experiment results show that the improved BP neural network model based on the building fire detection is strong in operation, which can be applied into the building fire detection, and it＇s worth for spreading use.