Performance analysis of Artificial Intelligent Techniques Based Rotor Position Estimation of SRM Drive [articol]

Abstract

Simple construction and low cost leads the Switched Reluctance Motor (SRM) applicable for many commercial applications. Rotor position sensing is necessary for the working of SRM. Rotor position sensors are expensive, occupies more space, not reliable, complex in connectivity, has maintenance issues and causes mechanical alignments problems. This paper focuses on the design of a sensorless scheme for rotor position sensing to obtain accuracy for the entire speed range of SRM drives is described using Fuzzy, Adaptive Neural Fuzzy Inference System (ANFIS). The rotor position angle is predicted by using the interaction between flux linkages and stator phase currents in terms of fuzzy base rule. This system comprises of time series prediction and knowledge-based algorithm to analyse the position sensing error. Using this model, the desired angle is estimated and predicted which minimises the sensing error. The simulated output results display the necessity of rotor position prediction in SRM drives. The comparative results of rotor position angle using fuzzy and ANFIS are shown in this paper.