主管：中华人民共和国应急管理部
主办：应急管理部天津消防研究所
ISSN 1009-0029  CN 12-1311/TU
消防科学与技术

消防科学与技术 ›› 2021, Vol. 40 ›› Issue (5): 725-729.

• • 上一篇    下一篇

GA优化BP神经网络在多参量火灾探测中的应用

郑皓天1，朱俊奇2，贾荣田3   

  1. 1. 安徽理工大学安全科学与工程学院，安徽淮南232001；2. 安徽理工大学经济与管理学院，安徽淮南232001；3. 东北大学资源与土木工程学院，辽宁沈阳110136
  • 出版日期:2021-05-15 发布日期:2021-05-15
  • 通讯作者: 朱俊奇(1984-)，男，安徽萧县人，安徽理工大学副教授，博士研究生。
  • 作者简介:郑皓天(1998-)，男，河南永城人，安徽理工大学安全科学与工程学院硕士研究生，安徽省淮南市山南新区泰丰大街168 号，232001。
  • 基金资助:
    国家自然科学基金资助项目(71971003)；国家社科基金重大项目（20ZDA084）

Application of genetic algorithm to optimize BP neural network in multi-parameter fire detection

ZHENG Hao-tian1, ZHU Jun-qi2, JIA Rong-tian3   

  1. 1. College of Safety Science and Engineering, Anhui University of Science and Technology, Anhui huainan 232001, China; 2. College of Economics and Management, Anhui University of Science and Technology, Anhui Huainan 232001, China; 3. College of Resources and Civil Engineering, Northeastern University, Liaoning Shenyang 110136, China
  • Online:2021-05-15 Published:2021-05-15

摘要： 针对BP 神经网络的随机权重和阈值稳定性不高的问题，运用遗传算法(GA)对BP 神经网络的初始权重和阈值进行优化，提出了一种基于GA 优化BP 神经网络的多参量数据融合方法以实现火灾探测，提高火灾探测准确率和模型泛化性能，并利用该模型对标准明火和阴燃火中的温度、烟雾浓度和CO 浓度进行数据融合实现火灾探测。研究显示，相较单纯BP 神经网络，经GA 优化的BP 神经网络火灾探测算法能够更快速精确地实现火灾探测，探测精度有显著改善，火灾识别准确率提高至98.84%。

关键词: 火灾探测, 遗传算法, BP 神经网络, 多参量

Abstract: This article proposes a multi- parameter data fusion method based on genetic algorithm to optimize BP neural network to realize fire detection, which can significantly improve the accuracy of fire detection. In view of the low stability of the random weight and threshold of BP neural network, it is proposed to use genetic algorithm to optimize the BP neural network to optimize the initial weight and threshold of the neural network to improve the generalization performance of the model, and use the model to compare the standard data fusion of temperature, smoke concentration and CO concentration in open flame and smoldering fire realize fire detection. The simulation results show that, compared with the pure BP neural network, the BP neural network fire detection algorithm optimized by the genetic algorithm can realize fire detection more quickly and accurately, the accuracy of fire detection is significantly improved, and the accuracy of fire recognition is increased to 98.84%.

Key words: fire detection, genetic algorithm, BP neural network, multi-parameter