Highlights A BP neural network optimized by genetic algorithm is developed. The hybrid model can improve forecast results. Forecast results in the low-latitude region need to be improved.

    Abstract A hybrid back propagation artificial neural network (ANN) with a genetic algorithm (GA) is built to predict 1-h ahead vertical total electron content (TEC) of single station in China. An ionospheric TEC forecast model was developed based on TEC data from four dual-frequency GPS stations of BJFS (39.61°N, 115.89°E), XIAN (34.18°N, 108.99°E), WUHN (30.53°N, 114.36°E), and KUNM (25.03°N, 102.80°E) in both 2007 and 2011. The results show the mean relative errors of the model are smaller than 10% and the root mean square (RMS) error is less than 1 TECU at mid-high-latitudes during the period of low solar activity in 2007. Some large RMS errors with more than 5 TECU can be observed at low latitude during the medium solar activity period of 2011. Compared to the traditional BP algorithm, the new algorithm is a promising and reliable alternative forecast technology for ionospheric TEC.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Forecasting of ionospheric vertical TEC 1-h ahead using a genetic algorithm and neural network


    Contributors:
    Huang, Z. (author) / Li, Q.B. (author) / Yuan, H. (author)

    Published in:

    Advances in Space Research ; 55 , 7 ; 1775-1783


    Publication date :

    2015-01-23


    Size :

    9 pages




    Type of media :

    Article (Journal)


    Type of material :

    Electronic Resource


    Language :

    English