Prediction of Total Electron Content (TEC) using Neural Network over Anomaly Crest Region Bhopal

In the present work neural network has been developed for prediction of Vertical Total Electron Content (VTEC) over Bhopal (23.2°N, 77.4°E & MLAT 14.2°N), an equatorial anomaly crest region. The study is based on VTEC data recorded by a multi-frequency multi-constellation GNSS receiver installed at during May 2016 to August 2017. Neural Network (NN) consists artificial neurons that are used learn patterns adjusting weights between input, hidden and output layers computational modelling. order obtain optimal number in layer, performance four systems have tested. Out which, fourth predict with 7 input study. layer factor affects trained networks. includes year, day hour day, geographic latitude, longitude, SSN IRI NmF2. We compared TEC obtained from (NN TEC), GPS derived (GPS TEC) IRI-2016 model (IRI analysis made. observed Root Mean Square Error (RMSE) decreases increase neurons. RMSE comes out be lowest (1.96 TECU), when 41.The predicted NN estimated same trend 01:00 07:00 UT. modelled values good agreement all seasons 0.99 correlation coefficient was calculated summer months, but always underestimates seasons. excellently 2017 putting credentials only 0.55–32.16% relative errors were them whereas it reaches up 93.86% TEC. Thus, result shows proposed can diurnal seasonal more accurate than region Bhopal. However, inclusion IRI-NmF2 as neuron increases performance.