Open AccessComparative assessment of two different model reference adaptive system schemes for speed‐sensorless control of induction motor drives
Author(s) -
Kumar Rakesh,
Das Sukanta,
Chattopadhyay Ajit Kumar
Publication year - 2016
Publication title -
iet electric power applications
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.815
H-Index - 97
eISSN - 1751-8679
pISSN - 1751-8660
DOI - 10.1049/iet-epa.2015.0121
Subject(s) - mras , control theory (sociology) , stator , induction motor , adaptive system , rotor (electric) , torque , stability (learning theory) , vector control , matlab , sensitivity (control systems) , computer science , control engineering , engineering , electronic engineering , physics , control (management) , artificial intelligence , mechanical engineering , voltage , machine learning , electrical engineering , thermodynamics , operating system
This study presents a comparison between a well‐known instantaneous reactive power ( Q )‐based model reference adaptive system (MRAS) ( Q ‐MRAS) and a relatively new X ‐MRAS used for the speed estimation of induction motor drive without speed sensor from the stability point of view. Stability aspects and sensitivity analysis of both the schemes to machine parameters variation are examined. The instantaneous and steady‐state values of Q are considered to develop the reference and adaptive models respectively for the former, whereas those of a fictitious quantity X (i.e. * × ) are considered in the said models for the later. Non‐computation of flux in these schemes leads to the satisfactory estimation of speed over a wide range maintaining the drive's stability in all the 4‐quadrants. Both the Q ‐ and X ‐MRAS are found to be dependent on the rotor resistance variation, whereas X ‐MRAS is also influenced by the stator resistance variation. A small signal stability study over a wide range of stator and rotor resistance variations is carried out to investigate the drive's stability in the low‐speed region for both the MRAS schemes. Matlab/Simulink is used to obtain the simulation results for the proposed study. Experimental results are obtained through a dSPACE‐1104‐based laboratory prototype.