Modeling and analysis of fof2 data using narx neural networks and wavelets

O. V. Mandrikova   ; Yu. A. Polozov

Vestnik KRAUNC. Fiziko-matematičeskie nauki, Tome 41 (2022) no. 4, pp. 137-146

Résumé

The need to detect anomalies is of particular relevance in the problems of geophysical monitoring, it requires ensuring the accuracy and efficiency of the method. The paper proposes an approach based on NARX neural networks for the problem of modeling foF2 data and detecting anomalies in them. It is known that neural networks are difficult to model highly noisy and essentially non- stationary time series. Therefore, the optimization of the process of modeling time series of a complex structure by the NARX network was performed using wavelet filtering. Using the example of processing time series of ionospheric parameters, the effectiveness of the proposed approach is shown, and the results for the problem of detecting ionospheric anomalies are presented. The approach can be applied when performing a space weather forecast to predict the parameters of the ionosphere.

Détail

Keywords: time series model, wavelet transform, neural network NARX, ionospheric parameters.

Bibliographie

[1] Alperovich L., Eppelbaum L., Zheludev V., Dumoulin J., Soldovieri F., Proto M., Bavusi M., Loperte A., “A New Combined Wavelet Methodology: Implementation to GPR and ERT Data Obtained in the Montagnole Experiment”, J. Geophys. Eng., 10 (2013), 025017 DOI: 10.1088/1742-2132/10/2/025017 | DOI

[2] Tang R., Zeng F., Chen Z., Wang J.-S., Huang C.-M., Wu Z., “The Comparison of Predicting Storm-Time Ionospheric TEC by Three Methods: ARIMA, LSTM, and Seq2Seq”, Atmosphere, 11 (2020), 316 DOI: 10.3390/atmos11040316 | DOI

[3] Kim S.-K., Lee E., Park J., Shin S., “TFeasibility Analysis of GNSS-Reflectometry for Monitoring Coastal Hazards”, Remote Sensing, 13 (2021), 976 DOI: 10.3390/rs13050976 | DOI

[4] Perol T., Gharbi M., Denolle M., “Convolutional Neural Network for Earthquake Detection and Location”, Sci. Adv., 4 (2018), e1700578 DOI: 10.1126/sciadv.1700578 | DOI

[5] Pavlicko M., Vojteková M., Blažeková O., “Forecasting of Electrical Energy Consumption in Slovakia”, Mathematics, 10 (2022), 577 DOI: 10.3390/math10040577 | DOI

[6] Ma Q., Liu S., Fan X., Chai C., Wang Y., Yang K., “A Time Series Prediction Model of Foundation Pit Deformation Based on Empirical Wavelet Transform and NARX Network”, Mathematics, 8 (2020), 1535 DOI: 10.3390/math8091535 | DOI

[7] Haykin S., Neural Networks: A Comprehensive Foundation, 2nd ed., Prentice Hall, N.J., 1999, 842 pp.

[8] Diaconescu E., “The use of NARX neural networks to predict chaotic time series”, WSEAS Trans. Comp. Res., 3 (2008), 182–191

[9] Gao Y., Er M.J., “NARMAX Time Series Model Prediction: Feedforward and Recurrent Fuzzy Neural Network Approaches”, Fuzzy Sets and Systems, 150 (2005), 331–350 DOI: 10.1016/j.fss.2004.09.015 | DOI

[10] Mallat S., A Wavelet Tour of Signal Processing, Academic Press, San Diego, 1999, 620 pp.

[11] Danilov A.D., “Ionospheric F-Region Response to Geomagnetic Disturbances”, Advances in Space Research, 52 (2013), 343–366 DOI: 10.1016/j.asr.2013.04.019 | DOI

[12] Mandrikova O., Mandrikova, B., “Method of wavelet-decomposition to research cosmic ray variations: Application in space weather”, Symmetry, 13:12 (2021), 2313 DOI: 10.3390/sym13122313 | DOI

[13] Mandrikova O., Mandrikova, B., “Hybrid method for detecting anomalies in cosmic ray variations using neural networks autoencoder”, Symmetry, 14:4 (2022), 744 DOI: 10.3390/sym14040744 | DOI

[14] Mandrikova O., Fetisova N., Polozov Y., “Hybrid model for time series of complex structure with ARIMA components”, Mathematics, 9:10 (2021), 1122 DOI: 10.3390/math9101122 | DOI