基于神经网络的经济圈道路交通安全综合测度模型

Compressive measurement of economic circles road traffic safety based on neural network

为提高交通安全评估的准确性、简化评估过程,以我国3大经济圈范围内道路交通安全状况为研究对象,运用神经网络理论构建模糊测度模型。在分析经济圈道路交通安全特性的基础上,建立风险程度、事故烈度、经济影响3层测度指标体系。利用模糊一致性判别矩阵,界定测度因子综合权重,确定测度区间阈值,将测度集分为严重危险、中度危险、基本危险、基本安全、安全等5个等级。以经济圈道路交通事故数据作为测度对象,利用神经网络对测度模型进行融合训练。结果表明,神经网络模型的训练正确率为100%,且能够有效简化评估过程。

In order to improve the accuracy of traffic safety assessment,and simplify the measurement process,the neural network theory was used to build a fuzzy model for evaluating economic circles road traffic safety. On the basis of analyzing the characteristics of road traffic safety in the economic circles,a three level measurement index system was established. By using a fuzzy consistency discriminate matrix,comprehensive weights of measurement factors were determined,and the threshold value of road traffic safety was determined to evaluate the safety situation. Five levels were set for the road traffic safety evaluation set. They are serious risk,moderate risk,basic risk,basic safety,and safety levels. The three main economic circles traffic accident data were taken as evaluation objects. The model was trained by using the neura network. Results show that the training accuracy of the model was 100%,and that the model can simplify effectively the measurement process.