ANN-Based DTC for Three level VSI Powered Induction Motor

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

This paper presents an artificial neural network (ANN) based direct torque control (DTC) drive for three-level VSI. Traditional look-up table (LUT)-based DTC algorithms have high output torque ripple and flux ripple. Additionally, CDTC is difficult to practically implement in multi-phase or multi-level systems due to exponential increases in memory requirements. The proposed ANN-based approach offers better generalization, reduced memory usage, and lower computational complexity. The presented ANN-based DTC eliminates hysteresis controllers, leading to improved control of torque and flux and better transient and steady-state response. The proposed ANN-based DTC is verified through simulations and experimental tests. Comparative results are presented.

Original languageBritish English
Title of host publicationIECON 2024 - 50th Annual Conference of the IEEE Industrial Electronics Society, Proceedings
PublisherIEEE Computer Society
ISBN (Electronic)9781665464543
DOIs
StatePublished - 2024
Event50th Annual Conference of the IEEE Industrial Electronics Society, IECON 2024 - Chicago, United States
Duration: 3 Nov 20246 Nov 2024

Publication series

NameIECON Proceedings (Industrial Electronics Conference)
ISSN (Print)2162-4704
ISSN (Electronic)2577-1647

Conference

Conference50th Annual Conference of the IEEE Industrial Electronics Society, IECON 2024
Country/TerritoryUnited States
CityChicago
Period3/11/246/11/24

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

  1. SDG 7 - Affordable and Clean Energy
    SDG 7 Affordable and Clean Energy

Keywords

  • Artificial neural network
  • direct torque control
  • electric vehicle
  • induction motor
  • voltage source inverter

Fingerprint

Dive into the research topics of 'ANN-Based DTC for Three level VSI Powered Induction Motor'. Together they form a unique fingerprint.

Cite this