基于神经网络和遗传算法的城市火灾风险评价模型

以消防安全工程学与系统安全工程理论为基础,结合我国城市发展特征及消防安全管理状况,建立了城市区域火灾风险评价指标体系;针对神经网络易陷入局部极小而引起评价指标权值分布不合理的缺陷,提出了基于神经网络和遗传算法的城市火灾风险评价模型,该模型以火灾发生的可能性以及灾后的严重程度为输入单元,火灾风险等级为输出单元,采用误差反算法训练BP网络,最终得出火灾风险等级范围,有效地解决了城市火灾的动态性和非线性特征;研究实例证明了该模型的有效性,可为城市的消防安全管理提供确实可行的参考依据。

The Model for Risk Evaluation of Urban Fire Based on Neural Network and Genetic Algorithms

According to fire control safety engineering and systematic safety engineering theory,risk evaluation index system for urban region fire is established based on the situation of cities' development of our country and the fire control safety management.Aiming at the irrational distribution of weight value of evaluation index caused by neural network's liability to local minimum,a new model for risk assessment of urban fire is established based on neural network and genetic algorithms.In this model,the likelihood of fire occurring and the severity caused by fire are regarded as input parameters and fire risk grade as output parameter.By adopting error-inverse arithmetic to train BP network,the risk grade range of fire is obtained,which effectively solves the dynamic and non-linear characteristics of urban fire.Illustration shows this model is feasible and can give a good reference to safety management of urban fire control.