Title: Artificial intelligence-based rotor position estimation for a 6/4 pole switched reluctance machine from phase inductance

Authors: D. Susitra; S. Paramasivam

Addresses: Faculty of Electrical and Electronics Engineering, Sathyabama University, Chennai 600 119, India ' ESAB India Limited, Equipment Factory, G-22, SIPCOT Industrial Park, Irungattukottai, Pennalur Post, Sriperumbudur Taluk, Kanchipuram District, 602 105, India

Abstract: This paper presents artificial intelligence-based rotor position estimation techniques for a three phase, 6/4 pole switched reluctance machine one based on fuzzy logic and another on adaptive neuro-fuzzy inference system (ANFIS). These techniques are applied for modelling the nonlinear rotor position of SRM using the magnetisation characteristics of the machine. Fuzzy logic technique is greatly suited to model general nonlinear mapping between input and output spaces. ANFIS has a strong nonlinear approximation ability which could be used for nonlinear modelling of the machine and its real time implementations. In this paper, the best features of fuzzy logic and ANFIS are utilised to develop the computationally efficient rotor position model θ(I, L) for SRM. Mathematical model for θ(I, L) using fuzzy inference system (FIS) and ANFIS has been successfully arrived, tested and presented for various values of phase currents (Iph) and phase inductance (L) of a nonlinear SRM. It is observed that both FIS and ANFIS are highly suitable for rotor position θ(I, L) modelling of SRM which is tested to be in good agreement with the training and checking of data used for modelling.

Keywords: rotor position estimation; modelling; fuzzy inference systems; FIS; adaptive neuro-fuzzy inference system; ANFIS; switched reluctance machine; SRM; phase inductance; artificial intelligence; intelligent estimation; fuzzy logic; neural networks.

DOI: 10.1504/IJMIC.2014.063877

International Journal of Modelling, Identification and Control, 2014 Vol.22 No.1, pp.68 - 79

Published online: 27 Sep 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article