UDC 551.524.1
CSCSTI 37.21
Russian Classification of Professions by Education 05.02.03
Russian Library and Bibliographic Classification 260
Russian Trade and Bibliographic Classification 6326
BISAK SCI042000 Earth Sciences / Meteorology & Climatology
A hybrid method for forecasting surface air temperature for the next day has been developed and implemented, which uses a fully-connected neural network combined with preprocessing of the input signal by the method of variational mode decomposition. The overall value of the mean absolute error for the entire forecast was 0.35 oC.
surface temperature, fully-connected neural network, forecast
1. Huang N. E., Shen Z., Long S. R., et al. The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis // Proceedings of the Royal Society of London A. 1998. V. 454. P. 903–995.
2. Torres, M.E., Colominas M.A., Schlotthauer G., Flandrin P. A complete ensemble empirical mode decomposition with adaptive noise // 2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). 2011. P. 4144 –4147.
3. Gilles J. Empirical Wavelet Transform // IEEE Transactions on Signal Processing. 2013. — V. 61(16). P. 3999-4010.
4. Dragomiretskiy K., Zosso D. Variational mode decomposition // IEEE Transactions on Signal Processing. 2013. V. 62(3). P. 531-544.
5. Sholle F. Glubokoe obuchenie na Python // SPb.: Piter, 2018. 400 s.
