Parameter Identification of Reaction Wheel Disturbance Model Based WLS-SVR Method

Abstract:

In order to predict the micro-vibration disturbance of reaction wheel, an improved experience model has been provided, which expresses the effect of the reaction wheels natural frequencies being the dominating factor in amplifying the disturbance. WLS-SVR algorithm is applied to identify the parameters of this model. The robust characteristic of this algorithm promises the global optimal estimation can be achieved. Experiment has been set up to measure the disturbance of the reaction wheel. Data achieved from the experiment are used to get the parameters of model and verify the identification results of WLS-SVR algorithm. The results show the improved model describes the properties of disturbance more precisely and the WLS-SVR algorithm can achieve accurately identification results.

• Periodical: Applied Mechanics and Materials (Volume 421)
• Pages: 144-149
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.421.144
• Online since: September 2013
• Authors: Yun Feng Wang, Wei Cheng, Ming Yi Xia, Shuang Wu
• Keywords: Disturbance Model, Micro Vibration, Natural Frequency, Reaction Wheel, WLS-SVR

References

[1] Eyerman C E, Shea J F. A systems engineering approach to disturbance minimization for spacecraft utilizing controlled structures technology MIT SERC Report (1990).

[2] Bialke B. High fidelity mathematical modeling of reaction wheel performance. Proceedings of the 21st Annual American Astronautical Society Guidance and Control Conference, Univelt Inc Publishing , San Diego, USA (1998), pp.483-493.

[3] Masterson R A, Miller D W. Development and validation of empirical and analytical reaction wheel disturbance models MIT SERL Report (1999).

[4] Sun Jie, Zhao Yang, Wang Benli. Development of precise disturbance model of reaction wheel assembly on a spacecraft. Journal of Harbin Institute of Technology, Vol. 38 (2006), pp.520-522.

[5] Zhang Pengfei, Cheng Wei Disturbance modeling and parameter identification of reaction wheel assembly on spacecraft. Journal of Beijing University of Aeronautics and Astronautics, Vol. 36 (2010).

[6] Vapnic V N. The nature of statistical learning theory. Springer Verlag, New York (1995).

[7] Mita A, Fujimoto A. Active detection of loosened bolts using ultrasonic waves and support vector machines. Proceedings of the 5th International Workshop on Structural Health Monitoring (2005).

[8] Zhang J, Sato T, Iai S, et al. A pattern recognition technique for structural identification using observed signals: linear case studies. Engineering Structures, Vol, 30-15(2008), pp.1439-1446.

DOI: 10.1016/j.engstruct.2007.08.006

[9] Rojo, Alvarez J L, Martinez , Ramon M, de Prado, Cumplido M, et al. Support vector method for robust ARMA system identification. IEEE Transactions on Signal Processing, Vol. 52 (2004), pp.155-164.

DOI: 10.1109/tsp.2003.820084

[10] Suykens J A K, Vandewalle J, de Moor B. Optimal control by least squares support vector machines . Neural Network, Vol. 14- 1 (2001), pp.23-35.

DOI: 10.1016/s0893-6080(00)00077-0

[11] Suykens J A K, de Brabanter J, Lukas L, et al. Weighted least squares support vector machines: robustness and sparse approximation. Neuro computing, Vol. 48 (2002), pp.85-105.

DOI: 10.1016/s0925-2312(01)00644-0

[12] Fu Zhichao, Cheng Wei, Xu Cheng. LS-SVM based Method for Modal Parameter Identification. ACTA AERONAUTICA ET ASTRONAUTICA SINICA, Vol. 30 (2009).