Modeling and Simulation-Based Evaluation of a Flyback Converter Using Classical and Learning-Based Control Methods Under Avionics-Oriented Power Conditions

Huseyin Kurnaz; Salih Baris Ozturk

doi:10.1109/RAST69551.2026.11672631

Modeling and Simulation-Based Evaluation of a Flyback Converter Using Classical and Learning-Based Control Methods Under Avionics-Oriented Power Conditions

Huseyin Kurnaz^*
, Salih Baris Ozturk

^*Corresponding author for this work

Department of Electrical Engineering

Department of Electronics and Communication Engineering
Istanbul Technical University
ASELSAN Inc.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

This paper compares a proportional-integral-derivative (PID) controller, an adaptive artificial neural network (ANN) controller, and a compact deep reinforcement learning (DRL) controller for a flyback-derived 28 V to 12 V converter. An averaged model and an approximate small-signal duty-to-output transfer function are used for controller design and interpretation of the results. The methods are evaluated under startup, line-voltage disturbance, load-step, and light-load conditions near the continuous-conduction-mode/discontinuous-conduction-mode (CCM/DCM) boundary using overshoot, settling time, steady-state error, output ripple, and integral of absolute error.

Original language	English
Title of host publication	RAST 2026 - 11th International Conference on Recent Advances in Air and Space Technologies, Conference Program
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9798319546371
DOIs	https://doi.org/10.1109/RAST69551.2026.11672631
Publication status	Published - 2026
Event	11th International Conference on Recent Advances in Air and Space Technologies, Conference Program, RAST 2026 - Istanbul, Turkey Duration: 13 May 2026 → 15 May 2026

Publication series

Name	RAST 2026 - 11th International Conference on Recent Advances in Air and Space Technologies, Conference Program

Conference

Conference	11th International Conference on Recent Advances in Air and Space Technologies, Conference Program, RAST 2026
Country/Territory	Turkey
City	Istanbul
Period	13/05/26 → 15/05/26

Bibliographical note

Publisher Copyright:
© 2026 IEEE.

Keywords

artificial neural network (ANN)
avionics power systems
deep reinforcement learning (DRL)
flyback converter
state-action-reward-state-action (SARSA)

Access to Document

10.1109/RAST69551.2026.11672631

Cite this

Kurnaz, H., & Ozturk, S. B. (2026). Modeling and Simulation-Based Evaluation of a Flyback Converter Using Classical and Learning-Based Control Methods Under Avionics-Oriented Power Conditions. In RAST 2026 - 11th International Conference on Recent Advances in Air and Space Technologies, Conference Program (RAST 2026 - 11th International Conference on Recent Advances in Air and Space Technologies, Conference Program). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/RAST69551.2026.11672631

Kurnaz, Huseyin ; Ozturk, Salih Baris. / Modeling and Simulation-Based Evaluation of a Flyback Converter Using Classical and Learning-Based Control Methods Under Avionics-Oriented Power Conditions. RAST 2026 - 11th International Conference on Recent Advances in Air and Space Technologies, Conference Program. Institute of Electrical and Electronics Engineers Inc., 2026. (RAST 2026 - 11th International Conference on Recent Advances in Air and Space Technologies, Conference Program).

@inproceedings{e8c3cfa8b5274a7db9ddcbd306e37b19,

title = "Modeling and Simulation-Based Evaluation of a Flyback Converter Using Classical and Learning-Based Control Methods Under Avionics-Oriented Power Conditions",

abstract = "This paper compares a proportional-integral-derivative (PID) controller, an adaptive artificial neural network (ANN) controller, and a compact deep reinforcement learning (DRL) controller for a flyback-derived 28 V to 12 V converter. An averaged model and an approximate small-signal duty-to-output transfer function are used for controller design and interpretation of the results. The methods are evaluated under startup, line-voltage disturbance, load-step, and light-load conditions near the continuous-conduction-mode/discontinuous-conduction-mode (CCM/DCM) boundary using overshoot, settling time, steady-state error, output ripple, and integral of absolute error.",

keywords = "artificial neural network (ANN), avionics power systems, deep reinforcement learning (DRL), flyback converter, state-action-reward-state-action (SARSA)",

author = "Huseyin Kurnaz and Ozturk, \{Salih Baris\}",

note = "Publisher Copyright: {\textcopyright} 2026 IEEE.; 11th International Conference on Recent Advances in Air and Space Technologies, Conference Program, RAST 2026 ; Conference date: 13-05-2026 Through 15-05-2026",

year = "2026",

doi = "10.1109/RAST69551.2026.11672631",

language = "English",

series = "RAST 2026 - 11th International Conference on Recent Advances in Air and Space Technologies, Conference Program",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "RAST 2026 - 11th International Conference on Recent Advances in Air and Space Technologies, Conference Program",

address = "United States",

}

Kurnaz, H & Ozturk, SB 2026, Modeling and Simulation-Based Evaluation of a Flyback Converter Using Classical and Learning-Based Control Methods Under Avionics-Oriented Power Conditions. in RAST 2026 - 11th International Conference on Recent Advances in Air and Space Technologies, Conference Program. RAST 2026 - 11th International Conference on Recent Advances in Air and Space Technologies, Conference Program, Institute of Electrical and Electronics Engineers Inc., 11th International Conference on Recent Advances in Air and Space Technologies, Conference Program, RAST 2026, Istanbul, Turkey, 13/05/26. https://doi.org/10.1109/RAST69551.2026.11672631

Modeling and Simulation-Based Evaluation of a Flyback Converter Using Classical and Learning-Based Control Methods Under Avionics-Oriented Power Conditions. / Kurnaz, Huseyin; Ozturk, Salih Baris.
RAST 2026 - 11th International Conference on Recent Advances in Air and Space Technologies, Conference Program. Institute of Electrical and Electronics Engineers Inc., 2026. (RAST 2026 - 11th International Conference on Recent Advances in Air and Space Technologies, Conference Program).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Modeling and Simulation-Based Evaluation of a Flyback Converter Using Classical and Learning-Based Control Methods Under Avionics-Oriented Power Conditions

AU - Kurnaz, Huseyin

AU - Ozturk, Salih Baris

PY - 2026

Y1 - 2026

N2 - This paper compares a proportional-integral-derivative (PID) controller, an adaptive artificial neural network (ANN) controller, and a compact deep reinforcement learning (DRL) controller for a flyback-derived 28 V to 12 V converter. An averaged model and an approximate small-signal duty-to-output transfer function are used for controller design and interpretation of the results. The methods are evaluated under startup, line-voltage disturbance, load-step, and light-load conditions near the continuous-conduction-mode/discontinuous-conduction-mode (CCM/DCM) boundary using overshoot, settling time, steady-state error, output ripple, and integral of absolute error.

AB - This paper compares a proportional-integral-derivative (PID) controller, an adaptive artificial neural network (ANN) controller, and a compact deep reinforcement learning (DRL) controller for a flyback-derived 28 V to 12 V converter. An averaged model and an approximate small-signal duty-to-output transfer function are used for controller design and interpretation of the results. The methods are evaluated under startup, line-voltage disturbance, load-step, and light-load conditions near the continuous-conduction-mode/discontinuous-conduction-mode (CCM/DCM) boundary using overshoot, settling time, steady-state error, output ripple, and integral of absolute error.

KW - artificial neural network (ANN)

KW - avionics power systems

KW - deep reinforcement learning (DRL)

KW - flyback converter

KW - state-action-reward-state-action (SARSA)

UR - https://www.scopus.com/pages/publications/105050710475

U2 - 10.1109/RAST69551.2026.11672631

DO - 10.1109/RAST69551.2026.11672631

M3 - Conference contribution

AN - SCOPUS:105050710475

T3 - RAST 2026 - 11th International Conference on Recent Advances in Air and Space Technologies, Conference Program

BT - RAST 2026 - 11th International Conference on Recent Advances in Air and Space Technologies, Conference Program

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 11th International Conference on Recent Advances in Air and Space Technologies, Conference Program, RAST 2026

Y2 - 13 May 2026 through 15 May 2026

ER -

Kurnaz H, Ozturk SB. Modeling and Simulation-Based Evaluation of a Flyback Converter Using Classical and Learning-Based Control Methods Under Avionics-Oriented Power Conditions. In RAST 2026 - 11th International Conference on Recent Advances in Air and Space Technologies, Conference Program. Institute of Electrical and Electronics Engineers Inc. 2026. (RAST 2026 - 11th International Conference on Recent Advances in Air and Space Technologies, Conference Program). doi: 10.1109/RAST69551.2026.11672631

Modeling and Simulation-Based Evaluation of a Flyback Converter Using Classical and Learning-Based Control Methods Under Avionics-Oriented Power Conditions

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this