Prediction of GNSS vertical coordinate time series is helpful to monitor crustal plate movement, dam or bridge deformation monitoring, global or regional coordinate system maintenance and so on. Machine learning is an effective method for time series prediction by manually inputting features, and its prediction results have strong interpretability. XGBoost algorithm is a machine learning algorithm that can evaluate features. It has good potential and stability for long-span time series prediction. By analyzing the applicability of XGBoost algorithm, a multi-station time series prediction model is proposed. Firstly, the model learns the characteristics of multi-station data in the data set, and then predicts the stations set as the target. The time series data of U direction from 2013 to 2015 at six stations such as BJYQ, BJSH and BJGB were selected for the experiment. The experimental results show that XGBoost algorithm has small prediction error and high prediction accuracy, and has strong generalization ability. It can be applied to GNSS time series prediction.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Prediction of Multistation GNSS Vertical Coordinate Time Series Based on XGBoost Algorithm


    Weitere Titelangaben:

    Lect. Notes Electrical Eng.


    Beteiligte:
    Yang, Changfeng (Herausgeber:in) / Xie, Jun (Herausgeber:in) / Li, Zhen (Autor:in) / Lu, Tieding (Autor:in)


    Erscheinungsdatum :

    2022-05-05


    Format / Umfang :

    12 pages





    Medientyp :

    Aufsatz/Kapitel (Buch)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch