基于BP神经网络的河南省火灾风险评价

采用火灾发生起数、死亡人数、受伤人数、直接经济损失、烧毁面积、受灾户数及人口火灾发生率7项指标构建河南省火灾风险评价综合体系,并用BP神经网络对该省份的18个地级市行政区的火灾风险进行评价,评价结果显示,该方法不但可以体现各地区的火灾风险相对水平,而且还可避免传统方法的主观性,具有较强的可行性和可靠性,对控制火灾风险、客观地反映各地区的消防工作现状具有一定的指导意义。

Risk Assessment of Fire Accidents in Henan Province of China Based on BP Neural Network

Seven indicators such as number of fire accidents, number of deaths,number of injuries,direct losses,burned area,number of household hit by fire and average number of fires per 100 thousand people are selected to construct fire risk assessment comprehensive system of Henan province.A BP(back-propagation) neural network was used to assess the risk level of 18 cities of Henan province.The result shows that this method can manifest relative fire risk level of those cities without subjectivity and is very practical and reliable.This study is very helpful to the control of fire risks and the objective reflection of fire-control situation in different regions.