基于连续贝叶斯网络的火灾预警模型研究  

Research on Fire Alarming Model Based on Continuous Bayesian Network

作　　者：吕灿[1] 张小波[1] 唐平[1] 

机构地区：[1]广东工业大学自动化学院,广东广州510006

出　　处：《工业控制计算机》2016年第2期104-106,共3页Industrial Control Computer

摘　　要：为了尽可能地减少森林火灾所带来的危害和损失,在火灾发生前进行火灾的预警预测是及其重要的。贝叶斯网络(Bayesian network),能够有效地处理不确定性事件,提出高斯贝叶斯网络结合模型,并将其应用到森林火灾的预防工作上,该模型不仅有助于构造网络结构,还可以处理连续变量。通过实验仿真,可知贝叶斯网络在森林火灾预警中,能够有效及时地预警火灾的发生,快速性较好。To minimize the damage and the loss of forest fires,early warning is importance in front of the fire.Bayesian Networks is an effective model for handling uncertainty.In this paper,Gaussian Bayesian network is proposed and used on the alarm of forest fire.This model not only is benefit for constructing the structure of the network,but also is worked on continuous variable. Bayesian network was used in the prevention of forest fires.

关 键 词：不确定性 贝叶斯网络 火灾预警 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]