Data-Driven Optimization for Neutral-Point-Clamped Dual-Active-Bridge Converter with Zero-Voltage-Zero-Current Switching

Jiaxin Dong; Xinze Li; Dhivya Sampath Kumar; Sivaneasan Bala Krishnan; Josep Pou; Anurag Sharma

doi:10.1109/IECON58223.2025.11221183

Data-Driven Optimization for Neutral-Point-Clamped Dual-Active-Bridge Converter with Zero-Voltage-Zero-Current Switching

Jiaxin Dong
, Xinze Li
, Dhivya Sampath Kumar
, Sivaneasan Bala Krishnan
, Josep Pou
, Anurag Sharma

Department of Electrical Engineering

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

Abstract

This paper presents an artificial intelligence (AI)-based optimization framework for improving the performance of a hybrid neutral-point-clamped dual-active-bridge (HNPCDAB) dc-dc converter under hybrid duty-ratio phase shift modulation (HDPSM). The proposed approach integrates simulation data from PLECS with the XGBoost machine learning algorithm to build predictive models for both soft-switching conditions and converter efficiency. These models are then used in conjunction with particle swarm optimization with state-based adaptive velocity limit strategy (PSO-SAVL) to identify optimal modulation parameters that achieve zero-voltage-zero-current switching (ZVZCS) while maximizing efficiency. Experimental validation on a 2-kW HNPCDAB prototype confirms the effectiveness of the proposed method across varying power levels and output voltages. Compared to conventional modulation strategies, the AI-optimized approach achieves improved soft-switching performance and higher efficiency, particularly under light-load conditions. © 2025 IEEE.

Original language	English
Title of host publication	IECON 2025 - 51st Annual Conference of the IEEE Industrial Electronics Society
Publisher	IEEE
ISBN (Electronic)	979-8-3315-9681-1
DOIs	https://doi.org/10.1109/IECON58223.2025.11221183
Publication status	Published - 2025
Event	51st Annual Conference of the IEEE Industrial Electronics Society (IECON 2025) - Madrid, Spain Duration: 14 Oct 2025 → 17 Oct 2025 https://iecon2025.org/

Publication series

Name	IECON Proceedings (Industrial Electronics Conference)
ISSN (Print)	2162-4704
ISSN (Electronic)	2577-1647

Conference

Conference	51st Annual Conference of the IEEE Industrial Electronics Society (IECON 2025)
Place	Spain
City	Madrid
Period	14/10/25 → 17/10/25
Internet address	https://iecon2025.org/

Funding

This work was supported by Energy Market Authority of Singapore under the EDGE Programme LA/Contract under Grant EDGE2-GC2022-008.

Research Keywords

Dual-active-bridge converter
neutral-point-clamped converter
zero-voltage-zero-current-switching

Access Output

10.1109/IECON58223.2025.11221183

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

Dong, J., Li, X., Kumar, D. S., Krishnan, S. B., Pou, J., & Sharma, A. (2025). Data-Driven Optimization for Neutral-Point-Clamped Dual-Active-Bridge Converter with Zero-Voltage-Zero-Current Switching. In IECON 2025 - 51st Annual Conference of the IEEE Industrial Electronics Society (IECON Proceedings (Industrial Electronics Conference)). IEEE. https://doi.org/10.1109/IECON58223.2025.11221183

@inproceedings{600916f368ef4560b285eccbe88ec0e5,

title = "Data-Driven Optimization for Neutral-Point-Clamped Dual-Active-Bridge Converter with Zero-Voltage-Zero-Current Switching",

abstract = "This paper presents an artificial intelligence (AI)-based optimization framework for improving the performance of a hybrid neutral-point-clamped dual-active-bridge (HNPCDAB) dc-dc converter under hybrid duty-ratio phase shift modulation (HDPSM). The proposed approach integrates simulation data from PLECS with the XGBoost machine learning algorithm to build predictive models for both soft-switching conditions and converter efficiency. These models are then used in conjunction with particle swarm optimization with state-based adaptive velocity limit strategy (PSO-SAVL) to identify optimal modulation parameters that achieve zero-voltage-zero-current switching (ZVZCS) while maximizing efficiency. Experimental validation on a 2-kW HNPCDAB prototype confirms the effectiveness of the proposed method across varying power levels and output voltages. Compared to conventional modulation strategies, the AI-optimized approach achieves improved soft-switching performance and higher efficiency, particularly under light-load conditions. {\textcopyright} 2025 IEEE.",

keywords = "Dual-active-bridge converter, neutral-point-clamped converter, zero-voltage-zero-current-switching",

author = "Jiaxin Dong and Xinze Li and Kumar, \{Dhivya Sampath\} and Krishnan, \{Sivaneasan Bala\} and Josep Pou and Anurag Sharma",

year = "2025",

doi = "10.1109/IECON58223.2025.11221183",

language = "English",

series = "IECON Proceedings (Industrial Electronics Conference)",

publisher = "IEEE",

booktitle = "IECON 2025 - 51st Annual Conference of the IEEE Industrial Electronics Society",

address = "United States",

note = "51st Annual Conference of the IEEE Industrial Electronics Society (IECON 2025) ; Conference date: 14-10-2025 Through 17-10-2025",

url = "https://iecon2025.org/",

}

Dong, J, Li, X, Kumar, DS, Krishnan, SB, Pou, J & Sharma, A 2025, Data-Driven Optimization for Neutral-Point-Clamped Dual-Active-Bridge Converter with Zero-Voltage-Zero-Current Switching. in IECON 2025 - 51st Annual Conference of the IEEE Industrial Electronics Society. IECON Proceedings (Industrial Electronics Conference), IEEE, 51st Annual Conference of the IEEE Industrial Electronics Society (IECON 2025), Madrid, Spain, 14/10/25. https://doi.org/10.1109/IECON58223.2025.11221183

Data-Driven Optimization for Neutral-Point-Clamped Dual-Active-Bridge Converter with Zero-Voltage-Zero-Current Switching. / Dong, Jiaxin; Li, Xinze; Kumar, Dhivya Sampath et al.
IECON 2025 - 51st Annual Conference of the IEEE Industrial Electronics Society. IEEE, 2025. (IECON Proceedings (Industrial Electronics Conference)).

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - Data-Driven Optimization for Neutral-Point-Clamped Dual-Active-Bridge Converter with Zero-Voltage-Zero-Current Switching

AU - Dong, Jiaxin

AU - Li, Xinze

AU - Kumar, Dhivya Sampath

AU - Krishnan, Sivaneasan Bala

AU - Pou, Josep

AU - Sharma, Anurag

PY - 2025

Y1 - 2025

N2 - This paper presents an artificial intelligence (AI)-based optimization framework for improving the performance of a hybrid neutral-point-clamped dual-active-bridge (HNPCDAB) dc-dc converter under hybrid duty-ratio phase shift modulation (HDPSM). The proposed approach integrates simulation data from PLECS with the XGBoost machine learning algorithm to build predictive models for both soft-switching conditions and converter efficiency. These models are then used in conjunction with particle swarm optimization with state-based adaptive velocity limit strategy (PSO-SAVL) to identify optimal modulation parameters that achieve zero-voltage-zero-current switching (ZVZCS) while maximizing efficiency. Experimental validation on a 2-kW HNPCDAB prototype confirms the effectiveness of the proposed method across varying power levels and output voltages. Compared to conventional modulation strategies, the AI-optimized approach achieves improved soft-switching performance and higher efficiency, particularly under light-load conditions. © 2025 IEEE.

AB - This paper presents an artificial intelligence (AI)-based optimization framework for improving the performance of a hybrid neutral-point-clamped dual-active-bridge (HNPCDAB) dc-dc converter under hybrid duty-ratio phase shift modulation (HDPSM). The proposed approach integrates simulation data from PLECS with the XGBoost machine learning algorithm to build predictive models for both soft-switching conditions and converter efficiency. These models are then used in conjunction with particle swarm optimization with state-based adaptive velocity limit strategy (PSO-SAVL) to identify optimal modulation parameters that achieve zero-voltage-zero-current switching (ZVZCS) while maximizing efficiency. Experimental validation on a 2-kW HNPCDAB prototype confirms the effectiveness of the proposed method across varying power levels and output voltages. Compared to conventional modulation strategies, the AI-optimized approach achieves improved soft-switching performance and higher efficiency, particularly under light-load conditions. © 2025 IEEE.

KW - Dual-active-bridge converter

KW - neutral-point-clamped converter

KW - zero-voltage-zero-current-switching

UR - https://www.scopus.com/pages/publications/105024701354

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-105024701354&origin=recordpage

U2 - 10.1109/IECON58223.2025.11221183

DO - 10.1109/IECON58223.2025.11221183

M3 - RGC 32 - Refereed conference paper (with host publication)

T3 - IECON Proceedings (Industrial Electronics Conference)

BT - IECON 2025 - 51st Annual Conference of the IEEE Industrial Electronics Society

PB - IEEE

T2 - 51st Annual Conference of the IEEE Industrial Electronics Society (IECON 2025)

Y2 - 14 October 2025 through 17 October 2025

ER -

Dong J, Li X, Kumar DS, Krishnan SB, Pou J, Sharma A. Data-Driven Optimization for Neutral-Point-Clamped Dual-Active-Bridge Converter with Zero-Voltage-Zero-Current Switching. In IECON 2025 - 51st Annual Conference of the IEEE Industrial Electronics Society. IEEE. 2025. (IECON Proceedings (Industrial Electronics Conference)). Epub 2025 Nov 6. doi: 10.1109/IECON58223.2025.11221183

Data-Driven Optimization for Neutral-Point-Clamped Dual-Active-Bridge Converter with Zero-Voltage-Zero-Current Switching

Abstract

Publication series

Conference

Funding

Research Keywords

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this