Switched Reluctance Machine Modeling through Multilayer Neural Networks A. C. F. Mamede, J. R. Camacho and R. E. Araújo   2018/04/20

Abstract

The work deals with the application of artificial neural networks (ANNs) in the modeling of switched reluctance machines (SRMs). The performance of a SRM is determined by its geometry, materials used and levels of excitation. In this way, this work investigates the influence of the stator and rotor back iron thickness in the performance of SRM. A multilayer neural network is proposed to learn the nonlinear characteristics of the motor. Data of flux linkages and torque are obtained through simulations of finite elements and used for ANN training. The algorithm developed in Octave allows the user to adjust the network parameters. The results presented confirm the feasibility of using ANN to establish a predictive model of SRM performance, thus enabling further investigation in the future.

Authors and affiliations

A. C. F. Mamede1, J. R. Camacho1 and R. E. Araújo2
1. Department of Electrical Engineering. Universidade Federal de Uberlândia (UFU). Campus Santa Mônic. Uberlândia (Brazil)
2. INESC TEC and Faculty of Engineering. University of Porto. (Portugal)

Key words

Switched reluctance machine, artificial neural networks, multilayer, modeling, predictive model, SRM performance.

References

[1] J. Ahn, "Switched reluctance motor", in Torque Control, 1st ed., M. Lamchich, Ed. Intech, 2011.
[2] R. Krishnan, Switched reluctance motor drives. Boca Raton, FL: CRC Press, 2001.
[3] P. Bajec, B. Zidaric and D. Miljavec, "Artificial Neural Network Modeling of Synchronous Reluctance Motor", in Progress In Electromagnetics Research Symposium, Marrakesh, 2011, pp. 1854-1858.
[4] Lu, W., Keyhani, A. and Fardoun, A. (2003). Neural network-based modeling and parameter identification of switched reluctance motors. IEEE Transactions on Energy Conversion, 18(2), pp.284-290.
[5] F. Kucuk, H. Goto, H. Guo and O. Ichinokura, "Artificial neural network based torque calculation of switched reluctance motor without locking the rotor", Journal of Applied Physics, vol. 105, no. 7, pp. 07F103, 2009.
[6] Ustun, O. (2009). A nonlinear full model of switched reluctance motor with artificial neural network. Energy Conversion and Management, 50(9), pp.2413-2421.
[7] Meeker, D. C. (2016). Finite Element Method Magnetics. [online] Version 4.2 (12Jan2016 Build). Available at: http://www.femm.info [Accessed 14 Nov. 2017].
[8] Athani, V. and Walivadekar, V. (1994). Equivalent Circuit for Switched Reluctance Motor. Electric Machines & Power Systems, 22(4), pp.533-543.
[9] L. Fausett, Fundamentals of neural networks, 1st ed. Upper Saddle River, N.J.: Prentice-Hall, 1994.
[10] S. Haykin and P. Engel, Redes neurais, 1st ed. Porto Alegre: Bookman, 2001.