The invention discloses a short-time traffic flow prediction method based on variational mode decomposition multi-stage optimization, which comprises the following steps: firstly, decomposing original traffic flow into a series of intrinsic mode components by adopting variational mode decomposition, respectively calculating sample entropy values of the intrinsic mode components, and merging the sample entropy values into trend, detail and random components; in order to further extract sequence information, each component is subjected to phase-space reconstruction. Normalizing the components and dividing the components into a training set and a test set; a least square support vector machine (LSSVM) is used to predict a trend component, a long short-term memory neural network (LSTM) is used to predict a detail component, and an extreme gradient boost (XGBoost) is used to predict a random component. And optimizing prediction model parameters by adopting an improved Australian wild dog algorithm (IDOA). And adding the outputs of the three prediction models to obtain a prediction value. According to the short-time traffic flow prediction model based on variational mode decomposition multi-stage optimization, the problem of low prediction precision caused by nonlinearity and randomness of original data is solved, and the prediction accuracy is improved.

    本发明公开了一种基于变分模态分解多阶段优化的短时交通流预测方法,首先采用变分模态分解将原始的交通流量分解为一系列的固有模态分量,分别计算其样本熵值并将其合并为趋势,细节和随机分量。为进一步提取序列信息,对各分量进行相空间重构。将各分量进行归一化并划分为训练集和测试集。使用最小二乘支持向量机(LSSVM)预测趋势分量,长短期记忆神经网络(LSTM)预测细节分量,极端梯度提升(XGBoost)预测随机分量。采用改进澳洲野狗算法(IDOA)优化预测模型参数。将三个预测模型的输出相加得到预测值。本发明构建的基于变分模态分解多阶段优化的短时交通流预测模型降低了由于原始数据非线性和随机性引起的预测精度不高的问题,提高了预测的准确度。


    Access

    Download


    Export, share and cite



    Title :

    Short-term traffic flow prediction model based on variational mode decomposition multi-stage optimization


    Additional title:

    基于变分模态分解多阶段优化的短时交通流预测模型


    Contributors:
    CHEN YI (author) / QI XINGYU (author) / HU SHUIYUAN (author) / YAO YUCHEN (author)

    Publication date :

    2023-09-08


    Type of media :

    Patent


    Type of material :

    Electronic Resource


    Language :

    Chinese


    Classification :

    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



    A Low Rank Dynamic Mode Decomposition Model for Short-Term Traffic Flow Prediction

    Yu, Yadong / Zhang, Yong / Qian, Sean et al. | IEEE | 2021


    Short-term traffic flow prediction method based on three-stage model

    KONG DEHUI / QIU PENGFEI / WANG SHAOFAN et al. | European Patent Office | 2021

    Free access



    Event-Based Short-Term Traffic Flow Prediction Model

    Head, K. L. / National Research Council / Transportation Research Board | British Library Conference Proceedings | 1995