基于人工神经网络的城市火灾事故的预测方法

随着社会经济的飞速发展,城市化进程的加快和人口的迅速增长,城市火灾频繁发生,造成的损失呈上升趋势。针对城市火灾事故发生的特点,根据人工神经网络基本原理和特性,建立了城市火灾事故神经网络预测模型;为了更精确预测城市火灾事故的发生,将城市火灾事故分为了高峰期(春节)和非高峰期两个时段分别进行预测;应用神经网络预测模型和分时段相结合方法对某城市火灾事故进行了实际预测。结果表明,神经网络模型是城市火灾事故预测的有效工具,该模型与时段法的结合能准确预测火灾事故发生的趋势。

Prediction Method of City Fire Accidents Based on Artificial Neural Network

With the rapid development of economy, urbanization and population increasing, the frequency of city fire and the economic loss are rising. According to the characteristics of city fire, a prediction model of city fire accidents by applying basic principles and features of neural network is established. In order to predict city fire accidents more accurately, the city fire accidents are classified into the peak period (during spring festival) and the non-peak period. The method is exemplified in a city successfully. The result shows that the model could be taken as an effective tool to predict city fire accidents and could accurately predict the occurrence tendency of city fire accidents when combined with the time period classification.