A watercourse type reservoir ship navigation prediction method based on multivariate data analysis comprises the steps that a data period is divided, the lowest water level serves as the starting point of the period, the ship navigation upper limit is in the rising trend at the starting point of the period, and ship navigation data in the rising interval can be predicted in a self-attention mode; a navigation condition fitted by the self-attention algorithm takes an average value of historical similar data as a reference, so that the error rate is smaller, and the obtained prediction data better meets actual requirements; a self-attention prediction mode is small in calculated amount, and a prediction result is corrected by adopting a multiple-prediction mode under the condition that current period data is insufficient, so that the prediction result of a water level rising interval is updated in real time. When the water level in the cycle begins to drop, the data quantity of the current cycle becomes more, the predictability becomes higher, the neural network model is adopted for navigation prediction, and the neural network model only aims at the water recession time period in the cycle, so that the time duration needing to be predicted is greatly shortened, and the advantage of high prediction accuracy of the neural network can be further highlighted.

    一种基于多元数据分析的河道型水库船舶通航预测方法,通过划分数据周期,以最低水位为周期的起始点,在周期开始点,船舶通航上限呈上升趋势,可采用自注意力的方式对上升区间的船舶通航数据进行预测,自注意力算法拟合出的通航情况以历史相似数据的平均值作为参考,出错率更小得到的预测数据更满足实际需求;自注意力预测方式计算量小,在当前周期数据不足的情况下采用多次预测的方式来修正预测结果,使水位上升区间的预测结果实时更新。当周期之中水位开始下降时,当前周期的数据量变多,可预测性变强,采用神经网络模型进行通航预测,由于神经网络模型仅针对周期中的退水时段,大大缩短了需要预测的时间长度,可进一步凸显神经网络预测精确度高的优势。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    River type reservoir ship navigation prediction method based on multivariate data analysis


    Weitere Titelangaben:

    一种基于多元数据分析的河道型水库船舶通航预测方法


    Beteiligte:
    TIAN RUI (Autor:in) / GUO LE (Autor:in) / LIU YONGZHI (Autor:in) / SHU WEIMIN (Autor:in) / WANG XIANG (Autor:in) / GUAN TIESHENG (Autor:in) / FENG ZHIZHOU (Autor:in) / LIU XIAOYANG (Autor:in) / ZHAO JIANHUA (Autor:in) / ZOU HONGMEI (Autor:in)

    Erscheinungsdatum :

    2024-01-26


    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



    Inland river intelligent ship navigation encounter time prediction method

    WANG ZHIYUAN / HE WEI / WU YONG et al. | Europäisches Patentamt | 2024

    Freier Zugriff

    Navigation vector based ship maneuvering prediction

    Perera, Lokukaluge Prasad | BASE | 2017

    Freier Zugriff

    Ship navigation method, navigation system, and ship

    HIRASAWA HIROAKI / HIRAKAWA SHINICHI / MIYASHITA TETSUJI et al. | Europäisches Patentamt | 2022

    Freier Zugriff

    Inland river ship electric propulsion system and inland river ship navigation control method comprising same

    WANG FANG / ZHANG GANGSHE / ZHANG JIAN et al. | Europäisches Patentamt | 2021

    Freier Zugriff

    SHIP, SHIP NAVIGATION METHOD

    SHUTO YUTA / YAMATO KUNIAKI | Europäisches Patentamt | 2020

    Freier Zugriff