Open AccessReduced switch multilevel inverter for performance enhancement of induction motor drive with intelligent rotor resistance estimator
Author(s) -
Chitra A.,
Himavathi S.
Publication year - 2015
Publication title -
iet power electronics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.637
H-Index - 77
eISSN - 1755-4543
pISSN - 1755-4535
DOI - 10.1049/iet-pel.2014.0648
Subject(s) - estimator , pulse width modulation , inverter , control theory (sociology) , induction motor , electronic engineering , total harmonic distortion , digital signal processor , network topology , computer science , engineering , motor drive , voltage , digital signal processing , electrical engineering , mathematics , artificial intelligence , statistics , control (management) , mechanical engineering , operating system
Multilevel inverters (MLI) are suitable for renewable energy sources as they operate at lower switching frequency and voltage. The sinusoidal PWM (SPWM) techniques are popularly employed in MLIs to improve the drive performance. However as the number of switches increase it leads to more switching losses, higher price and size of the inverter. In this study, a modified reduced switch MLI (RSMLI) with five switches is proposed. Various multilevel inverter topologies with different multicarrier SPWM techniques such as phase disposition, phase opposition disposition, and alternative phase opposition disposition (APOD) are derived, analysed and presented for the proposed RSMLI. All the inverter topologies are simulated and the scheme with lowest THD is identified. The hardware setup is built for the proposed RSMLI using F28335 digital signal processor (DSP) for pulse generation and the results are validated. The vector control scheme of industrial standard requires an accurate and fast R r estimator for stable performance. Neural based R r estimator proposed by the authors is implemented on FPGA. The proposed RSMLI fed induction motor drive is examined with and without the neural based R r estimator. Encouraging results obtained using the proposed RSMLI are reported.