Accelerating atmosphere modeling: neural network enhancements for faster NRLMSISE calculations

• Autorzy: Kashyn Volodymyr , Choliy Vasyl

Identyfikatory

DOI

10.2478/arsa-2025-0007

• Języki publikacji: EN

Abstrakty

EN

NRLMSISE is an empirical model that allows us to predict temperatures and densities of the main atmospheric components. The model is widely used to evaluate atmospheric impacts on satellite orbits and laser beam refraction which come through the atmosphere, such as those used for Earth-satellite distance measurements. Model of the atmosphere is a valuable part of the Satellite Laser Ranging processing software like Kyiv Geodynamics (Juliette). Juliette is written in C++ and exploits the C++ clone of NRLMSISE written by the second author. The C++ version produces the same outputs as an official Fortran code. Accurate modeling of atmospheric influences on satellite motion requires performing numerous calculations along satellite orbits or laser beam paths, which are computationally intensive. By decreasing calculation time of NRLMSISE, we would not only save the modeling time but also give a prospect for a wider application of the model due to lowering computational resource demands. Our work demonstrates how the traditional NRLMSISE model can be effectively translated into a neural network. This conversion achieves significant performance gains on both CPU and GPU while maintaining acceptable accuracy when compared to the C++ implementation of NRLMSISE. We demonstrate the process of moving NRLMSISE to a neural network, the resulting accuracy, ease of running the trained model on CUDA-enabled GPUs, and the obtained boost of performance on both CPU and GPU.

Słowa kluczowe

EN

atmospheric modeling; NRLMSISE; neural networks; machine learning; performance

• Wydawca: Centrum Badań Kosmicznych PAN ; De Gruyter
• Czasopismo: Artificial Satellites : Journal of Planetary Geodesy
• Rocznik: 2025
• Tom: Vol. 60, No. 3
• Strony: 121--136
• Opis fizyczny: Bibliogr. 20 poz., rys., tab.

Twórcy

autor

Kashyn Volodymyr

• Main Astronomical Observatory, Kyiv, Ukraine

autor

Choliy Vasyl

• Main Astronomical Observatory, Kyiv, Ukraine

Bibliografia

• Choliy V. (2007). New GNSS processor (Juliette) for geodynamic and atmospheric tasks. Geophysical Research Abstracts, v.9, EGU General Assembly 2007.
• Cybenko, G. Approximation by superpositions of a sigmoidal function. Math. Control Signal Systems 2, 303–314 (1989). https://doi.org/10.1007/BF02551274.
• Elfwing, S., Uchibe, E. & Doya, K. (2018) ‘Sigmoid-weighted linear units for neural network function approximation in reinforcement learning’, Neural Networks, 107, pp. 3-11. doi: 10.1016/j.neunet.2017.12.012.
• Hampel, F.R., 1974. The influence curve and its role in robust estimation. Journal of the American Statistical Association, 69 (346), pp. 383-393. https://doi.org/10.1080/01621459.1974.10482962.
• Hornik, K., Stinchcombe, M. and White, H. (1989) ‘Multilayer feedforward networks are universal approximators’, Neural Networks, 2(5), pp. 359-366. https://doi.org/10.1016/0893- 6080(89)90020-8.
• Howard, A., Sandler, M., Chu, G., Chen, L.-C., Chen, B., Tan, M., Wang, W., Zhu, Y., Pang, R., Vasudevan, V., Le, Q. V., & Adam, H. (2019, November 20). Searching for MobileNetV3 [Preprint]. arXiv. https://arxiv.org/abs/1905.02244.
• Izzo, D., Acciarini, G. & Biscani, F. (2024) ‘NeuralODEs for VLEO simulations: Introducing thermoNET for Thermosphere Modeling’. Proceedings of the 29th International Symposium on Space Flight Dynamics (ISSFD 2024), Darmstadt, Germany, 22-26 April 2024. ESA/ESOC. Available at: https://issfd.org/ISSFD_2024/ISSFD2024_3-2.pdf (Accessed 16 July 2025).
• Khorrami, M.S. et al. (2023) ‘An artificial neural network for surrogate modeling of stress fields in viscoplastic polycrystalline materials’, npj Computational Materials, 9, 37. https://doi.org/10.1038/s41524-023-00991-z.
• Kingma, D.P. and Ba, J. (2015) ‘Adam: A method for stochastic optimization’, Proceedings of the 3rd International Conference on Learning Representations (ICLR 2015), San Diego, CA, USA, 7-9 May. Available at: https://arxiv.org/abs/1412.6980.
• Loshchilov, I. and Hutter, F. (2019) ‘Decoupled weight decay regularization’, Proceedings of the 7th International Conference on Learning Representations (ICLR 2019), New Orleans, LA, USA, 6-9 May. Available at: https://arxiv.org/abs/1711.05101.
• Matzka, J., Bronkalla, O., Tornow, K., Elger, K. and Stolle, C., 2021. Geomagnetic Kp index. V. 1.0. GFZ Data Services, https://doi.org/10.5880/Kp.0001.
• NASA/GSFC CCMC (2025) NRLMSISE-00 empirical atmosphere model, Fortran source code. Available at: https://ccmc.gsfc.nasa.gov/modelweb/models/nrlmsise00.php (Accessed: 5 July 2025).
• Picone, J. M., A. E. Hedin, D. P. Drob, and A. C. Aikin, NRLMSISE-00 empirical model of the atmosphere: Statistical comparisons and scientific issues, J. Geophys. Res., 107(A12), 1468, doi:10.1029/2002JA009430, 2002.
• Raina, R., Madhavan, A. & Ng, A.Y. (2009) ‘Large-scale deep unsupervised learning using graphics processors’, Proceedings of the 26th International Conference on Machine Learning, pp. 873-880.
• Smith, L.N. (2015) ‘Cyclical learning rates for training neural networks. arXiv preprint arXiv:1506.01186. Available at: https://arxiv.org/abs/1506.01186.
• Smith, L.N. (2017) ‘Cyclical learning rates for training neural networks’, Proceedings of the IEEE Winter Conference on Applications of Computer Vision (WACV 2017), Santa Rosa, CA, 24-31 March Piscataway, NJ: IEEE, pp. 464-472. doi: 10.1109/WACV.2017.58.
• Smith, L.N. (2018) ‘A disciplined approach to neural-network hyper-parameters: Part 1 - learning rate, batch size, momentum and weight decay’. arXiv preprint arXiv:1803.09820. Available at: https://arxiv.org/abs/1803.09820.
• Tapping, K.F. (2013) ‘The 10.7 cm solar radio flux (F10.7)’, Space Weather, 11 (7), pp. 394-406. doi: 10.1002/swe.20064.
• Zhang, Q., Zhang, J., Liang, L., Li, Z. & Zhang, T. (2021) ‘A deep-learning-based surrogate model for estimating the flux and power distribution solved by diffusion equation’, EPJ Web of Conferences, 247, 03013. https://doi.org/10.1051/epjconf/202124703013.
• Zhang, Y., Yu, J., Chen, J., & Sang, J. (2021). An Empirical Atmospheric Density Calibration Model Based on Long Short-Term Memory Neural Network. Atmosphere, 12(7), 925. https://doi.org/10.3390/atmos12070925.