【摘要】 目的构建广西道路交通事故BP人工神经网络预测模型,为研究广西道路交通事故提供新方法。方法在分析道路交通事故与人、车、路等因素关系的基础上,选取人口数、客运周转量、民用车辆拥有量和公里里程数作为输入变量,交通事故发生数作为输出变量,应用BP人工神经网络技术,对2010年广西道路交通发生数进行预测。结果 2010年广西交通事故预测数为4 562次,实际发现4 351次,预测值与实际值误差为4.85%,建立的模型拟合效果较好。结论 BP人工神经网络模型适用于广西交通事故数的预测,为交通部门进行交通事故预测研究提供新方法.更多还原

【Abstract】 Objective To construct the prediction model of road traffic accidents in Guangxi by BP neural network,and to provide a new method for studying road traffic accidents.Methods Based on the analysis of the relation between road traffic accidents and factors,including human,vehicle and road,the predicting model of road traffic accidents,which used population,passenger turnover,number of civilian vehicles and Km mileage as the input neurons and the road traffic accidents as the output neuron,was established by BP neural network to predict the road traffic accidents of Guangxi in 2010.Results The predicted value and actual one in 2010 for the road traffic accidents were 4 562 and 4 351,respectively,and the percentage of error was 4.85%.The fitting of the model established was more effective.Conclusion The predicting model established by BP neural network is suited for predicting road traffic accidents in Guangxi,and it has provided a new method for traffic department.更多还原