智能家居火灾预警系统建模与分析

Modeling and Analysis of Intelligent Home Fire Warning System

为了实现让智能家居系统能够对家居火灾进行早期预报并及时准确的报警,从而保护家居环境和人身财产安全,开发了一种智能家居火灾预警系统;在预警系统中采用BP神经网络建立火灾预测模型,把家居温度、烟雾浓度、一氧化碳气体浓度多火灾信息作为输入参数,明火发生概率、阴燃火发生概率、无火发生概率作为输出,同时结合使用LM算法以及遗传算法对火灾预测模型进行大幅度优化;实验表明,该预警系统预测精度较高,有效地改善了传统火灾预测智能化程度低的问题。

In order to realize intelligent home system early warning and timely and accurate alarm to protect the home environment and personal property safety,a smart home fire early warning system is developed.In the early warning system was established by using BP neural network prediction model of fire,the home furnishing temperature,smoke concentration,the concentration of carbon monoxide gasfire information as input parameters,fire occurrence probability,the probability of occurrence of smoldering fire and no fire occurrence probability as output,combined with the use of the LM algorithm and genetic algorithm for optimization of fire prediction model greatly.Experiments show that the prediction accuracy of the warning system is high,and it can effectively improve the low intelligence level of the traditional fire prediction.