Ömrünü Tamamlamış Yer Sabit Uyduların Boylam Hareketlerinin Yapay Sinir Ağları ile Modellenmesi

Öz

Bu çalışmada, yapay sinir ağları kullanılarak işletme ömrünü tamamlamış yer sabit yörünge uydularının boylam hareketleri incelenmiştir. Uydu yörünge hareketleri ve dinamiği içinde, uydu boylam hareketleri yapay sinir ağları ile modellenmiştir. Ömrünü tamamlamış altı uyduya ait veriler, veri tabanından alınmış, kapsamlı bir ön işlemeye tabi tutulmuş ve hem tek girişli hem üç girişli yapay sinir ağı eğitiminde kullanılmıştır. Modelleme sonunda ölçülen ve tahmin edilen sonuçlar arasındaki ortalama kare hata (MSE) 1.55x10-3 ve regresyon değeri 1 civarında olup tüm uydular için oldukça başarılı sonuçlar elde edilmiştir. Böylece yapay sinir ağları ile karmaşık yörünge dinamiğinin modellenebildiği görülmüştür. İşletme ömrünü tamamlamış uyduların boylam hareketlerinin yapay sinir ağları ile etkili bir biçimde modellenebildiği görülmektedir. Uydu operatörleri bu tip uyduların uzun vadeli yörünge hareketlerini önerilen yöntem ile tahmin edebilir ve tahminlerini bu konuda alacakları kararlar için destek bilgisi olarak kullanabilir. İlave olarak bu araştırma ömrünü tamamlamış uyduların hareketlerini hassas bir şekilde göstermekte bu durumda daha iyi görev planlaması yapmaya, kaynak optimizasyonuna ve uzay enkazlarının daha iyi yönetilme stratejilerinin geliştirilmesine imkân tanımaktadır.

Anahtar Kelimeler

• Ömrünü tamamlamış uydu
• yer sabit yörünge
• yapay sinir ağları
• boylam değişimi
• yörünge dinamiği

Modeling Longitudinal Evolution of Decommissioned Geostationary Satellites using Neural Networks

Öz

This study uses neural networks to explore the intricate longitudinal progression of decommissioned geostationary satellites. The goal is to model and predict satellites' longitudinal dynamics across time dimensions. Historical satellite longitude data undergoes thorough preprocessing to train time series neural networks in both single-input and 3-input configurations for all six decommissioned satellites, yielding comprehensive longitudinal behavior insights. Results reveal impressive outcomes: average Mean Squared Error (MSE) between predicted and measured longitudes is 1.55x10-3, with regression close to unity. This convergence implies a strong alignment between the neural network methodology employed and the intricate problem domain. These results accentuate the suitability and effectiveness of the chosen neural network approach in addressing the challenges posed by decommissioned geostationary satellite trajectory modeling. The study's implications span various fields. Insight into long-term orbital shifts aids in understanding satellite behaviors, enhancing trajectory predictions and decision-making in satellite management and space technology advancement. Additionally the research emphasizes the importance of accurate predictions about satellite behavior after decommissioning. This contributes to better mission planning, resource optimization, and more efficient strategies for dealing with space debris.

Anahtar Kelimeler

• Decommissioned satellites
• geostationary orbits
• neural networks
• longitudinal evolution
• orbit dynamics.

Kaynakça

1. Soop EM. Introduction to geostationary orbits 1993: ESA.
2. Oz I. Coverages stabilization of an inclined orbit communication satellite with two axis biases. Journal of the Faculty of Engineering and Architecture 2022; 38:1, pp. 219-230.
3. Oz I, Yilmaz UC. Determination of coverage oscillation for inclined communication satellite. Sakarya University Journal of Science 2020; 24(5), 973-983.
4. ITU Radiocommunication Sector: Regulations and procedures for space radio communication, Recommendation ITU-R S.1003-1, 2021.
5. Inter-Agency Space Debris Coordination Committee (IADC): IADC Space debris mitigation guidelines. 2007; Issue 3.0.
6. European Space Agency (ESA): Space debris mitigation handbook. ESA Bulletin, 2005; Issue 123.
7. Büyükkeçeci M, Okur MC. A comprehensive review of feature selection and feature selection stability in machine learning. Gazi University Journal of Science, 2024; 1-10.
8. Ameur T, Eddine A, Benalia A. ANN identification technique and fuzzy pi control of a hybrid indirect matrix converter with a flying capacitor three level inverter in power active filtering application. Gazi University Journal of Science, 2023;1-10.

9. Seymen OF. et al. Customer churn prediction using ordinary artificial neural network and convolutional neural network algorithms: a comparative performance assessment. Gazi University Journal of Science, 2023.
10. Erdem OE, Kesen SE. Estimation of Turkey’s natural gas consumption by machine learning techniques. Gazi University Journal of Science 2020; 33.1: 120-133.
11. Rayan, AB, Artuner H. LSTM-Based deep learning methods for prediction of earthquakes using ionospheric data. Gazi University Journal of Science 2022; 35.4: 1417-1431.
12. Wickramasinghe L, Ekanayake P, Jayasinghee J. Machine learning and statistical techniques for daily wind energy prediction. Gazi University Journal of Science 2022; 35.4: 1359-1370.
13. Alshari H, Odabas A. Machine learning model to diagnose diabetes type 2 based on health behavior. Gazi University Journal of Science 2022;35.3. 834-852.
14. Tombaloğlu B, Erdem H. Turkish speech recognition techniques and applications of recurrent units (LSTM and GRU). Gazi University Journal of Science 2021;34.4: 1035-1049.
15. Montenbruck O, Gill E. Satellite orbits: models, methods, and applications, Springer 2011.
16. Li HN. Geostationary satellite collocation, Springer 2010.
17. Piani S. Analytical model for propagation of debris clouds in geostationary orbit. Tesi di Laurea Magistrale in Space Engineering,Ingegneria Spaziale. 2022.
18. Thomas R, Linares R. A survey of longitudinal-shift maneuvers performed by geosynchronous satellites from 2010 to 2021, 2022; 73rd Astronautical Congress, Paris, France.
19. CarvalhoMoraes SJP, Prado A. Analysis of the orbital evolution of space debris using a solar sail and natural forces. Advances in Space Research 2022; v 70(1), 125-143.
20. Byoung-Sun L. East–West station-keeping maneuver strategy for COMS satellite using iterative process. Advances in space research 2011; 47.1: 149-159.
21. https://www.space-track.org/ last access March 2023.
22. Stepišnik T. Machine learning for effective spacecraft operation: Operating INTEGRAL through dynamic radiation environments. Advances in Space Research 2020; 69.11: 3909-3920.
23. Thomas R, Soleraa HE, Linaresa R. Geosynchronous satellite behavior classification via unsupervised machine learning. In 9th Space Traffic Management Conference. Austin, 2023; TX (Vol. 3).
24. Roberts T. Geosynchronous Satellite Maneuver Classification and Orbital Pattern Anomaly Detection via Supervised Machine Learning. Diss. Massachusetts Institute of Technology, 2021.
25. Solera HE, Linares R. Geosynchronous satellite pattern of life node detection and classification, 9th Space Traffic Management Conference, Austin TX, 2023.
26. Çelikel R. Gündoğdu A. ANN-based MPPT algorithm for photovoltaic systems. Turkish Journal of Science and Technology 202; 15.2: 101-110.
27. Toraman S, Turkoglu I. A new method for classifying colon cancer patients and healthy people from FTIR signals using wavelet transform and machine learning techniques. Journal of the Faculty of Engineering and Architecture of Gazi University 2020;35.2: 933-942.
28. Dogan F, Turkoglu I. Derin öğrenme modelleri ve uygulama alanlarına ilişkin bir derleme. Dicle Üniversitesi Mühendislik Fakültesi Mühendislik Dergisi 2019;10.2: 409-445.
29. Ariafar S, Rüdiger J. Long-term evolution of retired geostationary satellites. 4th European Conference on Space Debris, 2005; Vol. 587.
30. Baresi Net al. Long-term evolution of mid-altitude quasi-satellite orbits. Nonlinear dynamics, 2020; 99: 2743-2763.
31. Proietti S et al. Long-term orbit dynamics of decommissioned geostationary satellites. Acta Astronautica 2021;182: 559-573.