OAK

An Angle-Oriented Limiting Method for Modulated Model Predictive Control to Implement a Grid Simulator

Metadata Downloads
Abstract
Voltage emulator for power-hardware-in-the-loop test-bed needs fast dynamics,
and model predictive controller can provide that. Power electronics devices such as
inverter has relatively short computation time for control algorithm, reflecting finite
control set model predictive control is introduced. However, finite control set model
predictive control has important drawbacks. It has wide spread frequency spectrum
and large ripple even in the steady state. This paper discusses a modulated model
predictive control, which ensure fixed switching frequency using pulse width
modulation. When large reference changes, the modulated model predictive control
is expected to face constraints from voltage source inverters. The modulated model
predictive control is regarded as quadratic programming problem, but geometric
analysis is involved, it can be computed non-iterative manner for optimal control.
Futhermore, the predictive model is modeled reflecting zero order hold behavior in
stationary reference frame, and the cost function to improve transient response of
modulated model predictive control is proposed. Simulation results demonstrating
the proposed control algorithm are presented through step responses under different
load conditions and various reference values.
Author(s)
Wooyoung Kim
Issued Date
2024
Type
Thesis
URI
https://scholar.gist.ac.kr/handle/local/18871
공개 및 라이선스
  • 공개 구분공개
파일 목록
  • 관련 파일이 존재하지 않습니다.

Items in Repository are protected by copyright, with all rights reserved, unless otherwise indicated.