Neural network in predicting the Earth rotation
Authors: Gorbachevskaya A.P.  
Published in issue: #3(92)/2024  
DOI:  
Category: Physics  Chapter: Astrophysics 

Keywords: neural network, neural network approach, forecasting, Earth rotation, Earth orientation models, International Earth Rotation Service (IERS), recurrent neural networks 

Published: 17.07.2024 
The paper compares the International Earth Rotation Service (IERS) forecast with the forecasts obtained using a neural network. For this purpose, the pole position data obtained from the IERS source are used. The forecast results are being compared using the different recurrent neural network architectures in order to assess accuracy and efficiency of the neural network approach. The paper pays particular attention to analyzing differences between the IERS forecasts and the neural network forecasts, as well as to identifying the neural network probable advantages in this area. The paper concludes on the prospects for introducing a neural network in developing the more accurate Earth orientation models.
References
[1] Munk W., MacDonald G. The Earths Rotation. 1964, Moscow, Mir, 384 p.
[2] Moritz G., Müller A. The Earths Rotation: Theory and Observations. Kyiv, Naukova Dumka, 1992.
[3] Markov Yu.G., Perepelkin V.V., Sinitsyn I.N., Semendyaev N.N. Information Models of the Unevenness of the Earths Rotation. Inform. i ee primen., 2011, No. 5, pp. 17–35. (In Russ.).
[4] Yatskiv Ya.S., Mironov N.T., Korsun' A.A., Taradiy V.K. The Movement of the Poles and the Unevenness of the Earths Rotation. Moscow, VINITI, 1976, v. 12, part 1, 2. (In Russ.).
[5] Akulenko L.D., Markov Yu.G., Perepyolkin V.V. Irregularities of the Earths Rotation. DAN, 2007, v. 417, no. 4, pp. 483–488. (In Russ.).
[6] Akulenko L.D., Markov Yu.G., Perepyolkin V.V., Rykhlova L.V. Intraannual Irregularities of the Earth's Rotation. Astron. J., 2008, v. 85, no. 7, pp. 657–664. (In Russ.).
[7] IERS. Available at: https://www.iers.org/ (accessed November 25, 2023).
[8] Rojas R. Neural Networks: A Systematic Introduction. Heidelberg, Springer Berlin, 1996, 509 p.
[9] Charu A. Neural networks and deep learning. St. Petersburg, Dialectika, 2020, 752 p.(In Russ.).
[10] Kildishev G.S., Frenkel A.A. Time series analysis and forecasting. Moscow, Statistika, 1973.(In Russ.).