贝叶斯网络在火灾报警系统中的应用  

Bayesian Network's Application in Fire Alarm System

作　　者：陈静[1,2] 付敬奇[2] 

机构地区：[1]安徽理工大学电气学院,安徽淮南232001 [2]上海大学机电工程与自动化学院,上海200072

出　　处：《仪表技术》2011年第10期47-51,共5页Instrumentation Technology

基　　金：国家"863"计划基金资助项目(2007AA04Z174);淮南市科技计划项目(2009A05011)

摘　　要：在火灾报警系统中火灾概率分析存在不确定性因素问题,为此文章提出用贝叶斯网络对火灾概率进行分析。首先通过分析火灾燃烧原理,得到火灾概率与燃烧过程产生的物化特征之间的内在逻辑关系;在定义火灾燃烧特征参量作为贝叶斯网络节点变量的基础上,创建了基于Netica的火灾报警系统贝叶斯网络模型。通过概率推理和对节点的证据敏感性分析,验证了利用贝叶斯网络模型对火灾发生概率进行分析是可行的、有效的。Aimed at the problem of false alarms caused by information uncertainty existing in the fire alarm system,the multi-sensor Bayesian network is proposed to analyze the fire alarm system.The internal logic relationship between the fire alarm and the physical and chemical characteristics generated in the process of fire burning is obtained through analysis of the fire mechanism.Based on defining node variables in the Bayesian network,multi-sensor Bayesian network model for the fire alarm system is created in Netica.The node＇s sensitivity to findings is analyzed;probabilistic inference verified that analyzing fire probability through the multi-sensor Bayesian network model is feasible and effective.

关 键 词：贝叶斯网络 火灾报警系统 概率推理 

分 类 号：TP277[自动化与计算机技术—检测技术与自动化装置]