An ANN-Based Output-Error-Driven Incremental Model Predictive Control for Buck Converter Against Parameter Variations

Research output: Journal Publications and ReviewsRGC 21 - Publication in refereed journalpeer-review

4 Scopus Citations
View graph of relations

Author(s)

Detail(s)

Original languageEnglish
Pages (from-to)1230-1248
Number of pages19
Journal / PublicationIEEE Journal of Emerging and Selected Topics in Power Electronics
Volume12
Issue number2
Online published6 Mar 2023
Publication statusPublished - Apr 2024

Abstract

DC/DC buck converters have been widely applied in the power processing of distributed energy resources (DERs) penetrated DC microgrid. Model predictive control (MPC) provides optimal control for the buck converter, but it often requires multiple sensors or complex observers to realize zero-steady-state-error control against parameter variations. An output-error-driven incremental MPC (OEDIMPC) for buck converter is proposed in this paper. Different from the traditional concept of calculating the optimal duty cycle, the proposed MPC calculates the optimal rate of change of the duty cycle by using a reduced-state output-error-driven prediction model. As a result, the OEDIMPC achieves zero-steady-state-error control against parameter variations, including load value, input voltage, output inductor, and output capacitor, with significantly improved dynamic performance and minimal sensor and observer requirements Besides, small- and large-signal analyzes are used to study the stability boundary of the OEDIMPC under parameter variations, and a shallow layer artificial neural network (ANN) is utilized to realize the OEDIMPC on the hardware. The proposed technique has been evaluated experimentally on a buck converter with resistor load and constant power load (CPL), confirming its effectiveness. © 2023 IEEE.

Research Area(s)

  • artificial neural networks, buck converter, Buck converters, Capacitive sensors, Capacitors, DC microgrids, Explicit model predictive control, Inductors, Observers, parameter variation, Predictive models, sensorless, Voltage control, Artificial neural networks (ANNs), explicit model predictive control (EMPC)

Citation Format(s)

An ANN-Based Output-Error-Driven Incremental Model Predictive Control for Buck Converter Against Parameter Variations. / Xiang, Yangxiao; Chung, Henry Shu-Hung; Shen, Ruihua et al.
In: IEEE Journal of Emerging and Selected Topics in Power Electronics, Vol. 12, No. 2, 04.2024, p. 1230-1248.

Research output: Journal Publications and ReviewsRGC 21 - Publication in refereed journalpeer-review