The invention discloses a high-grade highway accident prediction method based on an Elman neural network, and the method comprises the following steps: collecting the pre-analyzed related data of the accident number, traffic volume and road alignment index of a high-grade highway in a mountainous area by taking a year as a unit; performing basic road section unit division on the pre-analyzed high-grade highway in the mountainous area based on the road geometric line shape index, and determining a basic road section unit division rule; based on a rough set theory, screening out geometric line shape indexes of the basic road section units which have prominent influence on the occurrence of the high-grade highway traffic accidents in the mountainous area; assigning a value to a geometric line shape index null value item of the basic road section unit with prominent influence; and establishing a high-grade highway accident prediction model based on the Ellmann neural network, predicting the traffic accidents of the high-grade highway basic section, and obtaining the traffic accident rate. According to the invention, prediction of high-grade highway traffic accidents is realized, the problem that high-grade highway accident prediction is difficult is solved, and a safety management basis is provided for high-grade highways.

    本发明公开了一种基于埃尔曼神经网络的高等级公路事故预测方法,包括以下步骤:以年为单位,收集预分析的山区高等级公路的事故数、交通量和道路线形指标相关数据;基于道路几何线形指标对预分析的山区高等级公路进行基本路段单元划分,确定基本路段单元划分规则;基于粗糙集理论筛选出对山区高等级公路交通事故发生有突出影响的基本路段单元的几何线形指标;对有突出影响的基本路段单元的几何线形指标空值项进行赋值;建立基于埃尔曼神经网络的高等级公路事故预测模型,对高等级公路基本路段交通事故进行预测,获取交通事故发生率。本发明实现高等级公路交通事故的预测,解决了高等级公路事故预测困难等问题,为高等级公路提供了安全管理依据。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    High-grade highway accident prediction method based on Ellmann neural network


    Weitere Titelangaben:

    一种基于埃尔曼神经网络的高等级公路事故预测方法


    Beteiligte:
    QIN WEI (Autor:in) / HUANG DEGENG (Autor:in) / WANG CHANGHAI (Autor:in) / TAN YANCHUN (Autor:in) / OU JIANCONG (Autor:in) / LIN JING (Autor:in) / LUO HAIYU (Autor:in) / ZHOU MINLU (Autor:in) / HU WENJUN (Autor:in) / YAO XITONG (Autor:in)

    Erscheinungsdatum :

    2022-10-11


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

    IPC:    G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / G06Q Datenverarbeitungssysteme oder -verfahren, besonders angepasst an verwaltungstechnische, geschäftliche, finanzielle oder betriebswirtschaftliche Zwecke, sowie an geschäftsbezogene Überwachungs- oder Voraussagezwecke , DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES



    Accident prediction model for public highway-rail grade crossings

    Lu, Pan / Tolliver, Denver | Elsevier | 2016



    Highway Traffic Accident Detection Method with Fuzzy Neural Network

    Wang, Feng-yuan / Huang, Bing-xi / Qu, Da-yi et al. | ASCE | 2007


    Highway Traffic Accident Detection Method with Fuzzy Neural Network

    Wang, F.-y. / Huang, B.-x. / Qu, D.-y. et al. | British Library Conference Proceedings | 2007


    Decision Tree Approach to Accident Prediction for Highway–Rail Grade Crossings: Empirical Analysis

    Zheng, Zijian / Lu, Pan / Tolliver, Denver | Transportation Research Record | 2016