基于机器学习的电池系统双向交错DC/DC变换器稳定控制

IF 1.7 4区工程技术 Q3 ENGINEERING, ELECTRICAL & ELECTRONIC

IET Power Electronics Pub Date : 2025-01-16 DOI:10.1049/pel2.12847

Ran Li, Ruigang Wang, Wendong Feng, Tianhao Qie, Yulin Liu, Tyrone Fernando, Herbert HoChing Iu, Xinan Zhang

{"title":"基于机器学习的电池系统双向交错DC/DC变换器稳定控制","authors":"Ran Li, Ruigang Wang, Wendong Feng, Tianhao Qie, Yulin Liu, Tyrone Fernando, Herbert HoChing Iu, Xinan Zhang","doi":"10.1049/pel2.12847","DOIUrl":null,"url":null,"abstract":"<p>This paper proposes an innovative machine learning-based control algorithm for the bidirectional interleaved DC/DC converter in battery systems. It offers fast dynamics, gain tuning-free control design, and guaranteed closed-loop stability. Furthermore, compared to the other learning-based control methods, the proposed algorithm shows low computational complexity in both the neural network training and the online digital implementation. The efficacy of the proposed method is substantiated through experimental validation.</p>","PeriodicalId":56302,"journal":{"name":"IET Power Electronics","volume":"18 1","pages":""},"PeriodicalIF":1.7000,"publicationDate":"2025-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1049/pel2.12847","citationCount":"0","resultStr":"{\"title\":\"A stable machine learning based control for bidirectional interleaved DC/DC converter in battery systems\",\"authors\":\"Ran Li, Ruigang Wang, Wendong Feng, Tianhao Qie, Yulin Liu, Tyrone Fernando, Herbert HoChing Iu, Xinan Zhang\",\"doi\":\"10.1049/pel2.12847\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p>This paper proposes an innovative machine learning-based control algorithm for the bidirectional interleaved DC/DC converter in battery systems. It offers fast dynamics, gain tuning-free control design, and guaranteed closed-loop stability. Furthermore, compared to the other learning-based control methods, the proposed algorithm shows low computational complexity in both the neural network training and the online digital implementation. The efficacy of the proposed method is substantiated through experimental validation.</p>\",\"PeriodicalId\":56302,\"journal\":{\"name\":\"IET Power Electronics\",\"volume\":\"18 1\",\"pages\":\"\"},\"PeriodicalIF\":1.7000,\"publicationDate\":\"2025-01-16\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://onlinelibrary.wiley.com/doi/epdf/10.1049/pel2.12847\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"IET Power Electronics\",\"FirstCategoryId\":\"5\",\"ListUrlMain\":\"https://onlinelibrary.wiley.com/doi/10.1049/pel2.12847\",\"RegionNum\":4,\"RegionCategory\":\"工程技术\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"ENGINEERING, ELECTRICAL & ELECTRONIC\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"IET Power Electronics","FirstCategoryId":"5","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1049/pel2.12847","RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ENGINEERING, ELECTRICAL & ELECTRONIC","Score":null,"Total":0}

引用次数: 0

摘要

本文提出了一种基于机器学习的电池系统双向交错DC/DC变换器控制算法。它提供快速动态，增益无调谐控制设计，并保证闭环稳定性。此外，与其他基于学习的控制方法相比，该算法在神经网络训练和在线数字实现方面都具有较低的计算复杂度。实验验证了该方法的有效性。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

A stable machine learning based control for bidirectional interleaved DC/DC converter in battery systems

查看原文本刊更多论文

A stable machine learning based control for bidirectional interleaved DC/DC converter in battery systems

This paper proposes an innovative machine learning-based control algorithm for the bidirectional interleaved DC/DC converter in battery systems. It offers fast dynamics, gain tuning-free control design, and guaranteed closed-loop stability. Furthermore, compared to the other learning-based control methods, the proposed algorithm shows low computational complexity in both the neural network training and the online digital implementation. The efficacy of the proposed method is substantiated through experimental validation.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

IET Power Electronics ENGINEERING, ELECTRICAL & ELECTRONIC-

CiteScore

5.50

自引率

10.00%

发文量

195

审稿时长

5.1 months

期刊介绍： IET Power Electronics aims to attract original research papers, short communications, review articles and power electronics related educational studies. The scope covers applications and technologies in the field of power electronics with special focus on cost-effective, efficient, power dense, environmental friendly and robust solutions, which includes: Applications: Electric drives/generators, renewable energy, industrial and consumable applications (including lighting, welding, heating, sub-sea applications, drilling and others), medical and military apparatus, utility applications, transport and space application, energy harvesting, telecommunications, energy storage management systems, home appliances. Technologies: Circuits: all type of converter topologies for low and high power applications including but not limited to: inverter, rectifier, dc/dc converter, power supplies, UPS, ac/ac converter, resonant converter, high frequency converter, hybrid converter, multilevel converter, power factor correction circuits and other advanced topologies. Components and Materials: switching devices and their control, inductors, sensors, transformers, capacitors, resistors, thermal management, filters, fuses and protection elements and other novel low-cost efficient components/materials. Control: techniques for controlling, analysing, modelling and/or simulation of power electronics circuits and complete power electronics systems. Design/Manufacturing/Testing: new multi-domain modelling, assembling and packaging technologies, advanced testing techniques. Environmental Impact: Electromagnetic Interference (EMI) reduction techniques, Electromagnetic Compatibility (EMC), limiting acoustic noise and vibration, recycling techniques, use of non-rare material. Education: teaching methods, programme and course design, use of technology in power electronics teaching, virtual laboratory and e-learning and fields within the scope of interest. Special Issues. Current Call for papers: Harmonic Mitigation Techniques and Grid Robustness in Power Electronic-Based Power Systems - https://digital-library.theiet.org/files/IET_PEL_CFP_HMTGRPEPS.pdf