Model Predictive Control Algorithm for Prolonging Lifetime of Three-Phase Voltage Source Converters

One of the key goals of power electronics is to reduce losses in three-phase converters in order to increase efficiency and reduce thermal stress, which can lengthen the lifetime of devices. However, the aforementioned studies do not account for the fact that the phase legs of the converter can expe...

Full description

Bibliographic Details
Main Authors: Minh-Hoang Nguyen, sangshin kwak, Seungdeog Choi
Format: Article
Language:English
Published: IEEE 2023-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10182233/
Description
Summary:One of the key goals of power electronics is to reduce losses in three-phase converters in order to increase efficiency and reduce thermal stress, which can lengthen the lifetime of devices. However, the aforementioned studies do not account for the fact that the phase legs of the converter can experience various aging conditions, which reduces the expected lifetime of converter. This paper proposes a minimum loss per-phase method based on predictive control for voltage source inverter, which aims at particularly reducing loss of specific phase leg to increase the lifetime of the entire converter. The output currents will be controlled using predicted reference voltages and predictive zero-sequence voltage injection in the proposed technique. The particular phase leg, which is the most aged leg, will be clamped to the positive and negative dc-link to generate a non-switching region. This will result in the reduction of switching frequency and switching loss of the most aged phase leg in the voltage source inverter, which prolongs the lifetime of the particular phase leg and entire converter. The results of simulation and experimentation are used to confirm the validity of the proposed method and effectiveness in increasing lifetime of converter, where the switching loss of the most aged leg is reduced by up to 85%, and corresponding lifetime increases by about four times compared to the conventional model predictive control method.
ISSN:2169-3536