Skip to main content

Research Repository

Advanced Search

A Current Sensorless Computationally Efficient Model Predictive Control for Matrix Converters

Sarajian, Ali; Guan, Quanxue; Wheeler, Patrick; Khaburi, Davood Arab; Kennel, Ralph; Rodriquez, Jose

A Current Sensorless Computationally Efficient Model Predictive Control for Matrix Converters Thumbnail


Authors

Ali Sarajian

Quanxue Guan

Profile Image

PATRICK WHEELER pat.wheeler@nottingham.ac.uk
Professor of Power Electronic Systems

Davood Arab Khaburi

Ralph Kennel

Jose Rodriquez



Abstract

Model Predictive Control (MPC) is becoming more popular than ever as an alternative to conventional modulations such as Space Vector Modulation methods to control matrix converters (MCs). However, the implementation of MPC is computationally expensive, because control objectives are required to evaluate all admissible switching states of the converter. Additionally, a large number of sensors to measure the 3-phase load currents, source currents, source voltages, and input voltages of MCs increases the overall cost. To sort this out, an efficient MPC is proposed for MCs to enable fast computation and low cost. This approach eliminates the calculations of future load currents and source currents for all possible switching states, requiring only two predictions for the calculation of output voltage and input current references. Further, it removes all current sensors by employing a Luenberger observer. A simulation study has demonstrated that the proposed method can reduce the computation overhead and hardware cost dramatically, leading to high-frequency operation and good converter performance.

Citation

Sarajian, A., Guan, Q., Wheeler, P., Khaburi, D. A., Kennel, R., & Rodriquez, J. (2022). A Current Sensorless Computationally Efficient Model Predictive Control for Matrix Converters. In 48th Annual Conference of the Industrial Electronics Society (IECON 2022). https://doi.org/10.1109/iecon49645.2022.9969053

Conference Name IECON 2022 – 48th Annual Conference of the IEEE Industrial Electronics Society
Conference Location Brussels, Belgium
Start Date Oct 17, 2022
End Date Oct 20, 2022
Acceptance Date Jul 10, 2022
Online Publication Date Oct 17, 2022
Publication Date Oct 17, 2022
Deposit Date Sep 5, 2022
Publicly Available Date Mar 28, 2024
Publisher IEEE
Series Title Annual Conference of the Industrial Electronics Society
Series ISSN 1553-572X
Book Title 48th Annual Conference of the Industrial Electronics Society (IECON 2022)
ISBN 9781665480260
DOI https://doi.org/10.1109/iecon49645.2022.9969053
Public URL https://nottingham-repository.worktribe.com/output/10908813
Publisher URL https://ieeexplore.ieee.org/document/9969053
Related Public URLs https://iecon2022.org/

Files




You might also like



Downloadable Citations