论文部分内容阅读
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 assessment process, taking the road traffic safety conditions in the three major economic circles of our country as the research object, the neural network theory is used to construct the fuzzy measure model. Based on the analysis of the road traffic safety features in the economic circle, a three-level measure indicator system of risk degree, accident intensity and economic impact is established. The fuzzy consistency discrimination matrix is used to define the comprehensive weight of the measure factors and the threshold of the measure interval is determined. The measure set is divided into five levels: serious danger, moderate danger, basic danger, basic safety and security. Taking the data of road traffic accidents in the economic circle as the measuring object, the neural network is used to fuse the measurement model. The results show that the neural network model training accuracy is 100%, and can effectively simplify the assessment process.