SRM drives are the upcoming drives nowadays as these have many advantages such as simplicity , low manufacturing and operating costs, fault tolerance, high torque/inertia ratio and efficiency. The estimation of SRM drive parameters is an important consideration in their field. Many methods are available for this. However the estimation of the optimal parameters is normally preferred. Making use of neural networks is one of the best ways to achieve this. This paper proposes an unsupervised learning method i.e., Kohonen’s Self Organizing Feature Map method of estimation of SRM drives. Since the method makes use of ‘winner takes all’ of a neuron, the values obtained by this, will be the optimal values. The drive is first simulated and the parameters obtained are used for training the ANN. The Unsupervised learning method is the Kohonen’s Self Organizing Feature Map method, which is used for the estimation of the SRM drive parameters. The parameters estimated are the currents and fluxes in the two axis . Because of the unsupervised learning, it can be stated that the estimated values are the best or the optimal values. MATLAB/Simulink is used for the simulation and the results are shown.

1. K.M. Rahman, S. Gopalakrishnan, Optimized Instantaneous Torque Control of Switched Reluctance Motor by Neural Network, IEEE IAS, Vol. 37, 2001, pp. 904-913.
2. E. Mese, D. A. Torrey, An approach for Sensor less Position Estimation for Switched Reluctance Motors Using Artificial Neural Networks, IEEE Trans. Power Electronics; Vol.17: No. l, 2002, pp. 66-75.
3. Wenzhe Lu, Ali Keyhani, and Abbas Fardoun. Neural Network-Based Modeling and Parameter identification of Switched Reluctance Motors, IEEE transactions on energy conversion. June 2003;vol. 18:no. 2.
4. S. Mir, I. Husain, and M. E. Elbuluk, “Switched reluctance motor modeling with on-line parameter Identification,” IEEE Trans. Ind. Applicat., vol. 34, July/Aug. 1998:pp. 776–783,
5. B. Fahimi, G. Suresh, J. Mahdavi, and M. Ehsani, “A new approach to model switched reluctance motor drive application to dynamic performance prediction, control and design,” in Power Electron. Specialists Conf.,vol. 2, 1998.
6. Garside, J.J.; Brown, R.H.; Arkadan,A.A.; Identification of switched reluctance motor states using application specific artificial neural networks, Industrial Electronics, Control, and Instrumentation, 1995., Proceedings of the 1995 IEEE IECON 21st International Conference on, Volume: 2 , 6-10 Nov. 1995 Pages: 1446 -1451 vol.2
7. Elmas, C.; Sagiroglu, S.; Colak, I.; Bal, G.; Modelling of a nonlinear switched reluctance drive based on artificial neural networks Power Electronics and Variable-Speed Drives, 1994. Fifth International Conference on, 26-28 Oct 1994 Pages: 7 – 12
8. Reay, D.S.; Green, T.C.; Williams, B.W.; Minimisation of torque ripple in a switched reluctance motor using a neural network, Third International Conference on , 25-27 May 1993 Pages:224 – 228.

Artificial Neural Network d-q control Epoch Estimation KSOFM SRM Optimal Parameters Unsupervised Learning Unit Vectors Weight Matrix Epoch Estimation KSOFM SRM Optimal Parameters Unsupervised Learning Unit Vectors Weight Matrix