Energy Management Strategy for Fuel Cell Hybrid Electric Vehicle Based on Driving Pattern Recognition With Fuzzy C-Means Clustering

IF 7.1 1区工程技术 Q1 ENGINEERING, ELECTRICAL & ELECTRONIC

IEEE Transactions on Power Electronics Pub Date : 2026-08-01 Epub Date: 2026-03-12 DOI:10.1109/TPEL.2026.3673572

Zhanfeng Zhu;Zhendong Sun;Yingfang Liu;Zonghai Chen

{"title":"Energy Management Strategy for Fuel Cell Hybrid Electric Vehicle Based on Driving Pattern Recognition With Fuzzy C-Means Clustering","authors":"Zhanfeng Zhu;Zhendong Sun;Yingfang Liu;Zonghai Chen","doi":"10.1109/TPEL.2026.3673572","DOIUrl":null,"url":null,"abstract":"Fuel cell vehicles have broad prospects in the field of transportation electrification, while their operating characteristics limit their application in complex working conditions. As a result, fuel cell vehicles are generally equipped with Li-ion batteries or ultra-capacitors as auxiliary power sources. An energy management strategy (EMS) is essential to maintain the efficient, stable, and reliable operation of fuel cell hybrid electric vehicles. In this work, a driving pattern recognition (DPR) method is introduced to improve the adaptability of the EMS. First, fuzzy C-means clustering is applied to classify different driving patterns, facilitating the construction of representative driving cycles. Under each driving cycle, the parameters of fuzzy logic controllers are optimized using a genetic algorithm. Finally, the corresponding fuzzy logic controller is invoked based on the DPR results for power allocation in this hybrid power electronic system. The proposed method reduces the overall cost by at least 9.94% compared with a fuzzy logic control strategy, while remaining within 1.83% of the dynamic programming strategy, demonstrating its near-optimal performance with real-time feasibility.","PeriodicalId":13267,"journal":{"name":"IEEE Transactions on Power Electronics","volume":"41 8","pages":"13996-14006"},"PeriodicalIF":7.1000,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEEE Transactions on Power Electronics","FirstCategoryId":"5","ListUrlMain":"https://ieeexplore.ieee.org/document/11432985/","RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2026/3/12 0:00:00","PubModel":"Epub","JCR":"Q1","JCRName":"ENGINEERING, ELECTRICAL & ELECTRONIC","Score":null,"Total":0}

引用次数: 0

Abstract

Fuel cell vehicles have broad prospects in the field of transportation electrification, while their operating characteristics limit their application in complex working conditions. As a result, fuel cell vehicles are generally equipped with Li-ion batteries or ultra-capacitors as auxiliary power sources. An energy management strategy (EMS) is essential to maintain the efficient, stable, and reliable operation of fuel cell hybrid electric vehicles. In this work, a driving pattern recognition (DPR) method is introduced to improve the adaptability of the EMS. First, fuzzy C-means clustering is applied to classify different driving patterns, facilitating the construction of representative driving cycles. Under each driving cycle, the parameters of fuzzy logic controllers are optimized using a genetic algorithm. Finally, the corresponding fuzzy logic controller is invoked based on the DPR results for power allocation in this hybrid power electronic system. The proposed method reduces the overall cost by at least 9.94% compared with a fuzzy logic control strategy, while remaining within 1.83% of the dynamic programming strategy, demonstrating its near-optimal performance with real-time feasibility.

查看原文本刊更多论文

基于模糊c均值聚类驾驶模式识别的燃料电池混合动力汽车能量管理策略

燃料电池汽车在交通运输电气化领域具有广阔的应用前景，但其运行特性限制了其在复杂工况下的应用。因此，燃料电池汽车一般配备锂离子电池或超级电容器作为辅助电源。能源管理策略是保证燃料电池混合动力汽车高效、稳定、可靠运行的关键。本文引入了一种驱动模式识别（DPR）方法来提高EMS的自适应性。首先，采用模糊c均值聚类对不同的驾驶模式进行分类，便于构建具有代表性的驾驶工况；在每个工况下，采用遗传算法对模糊控制器参数进行优化。最后，根据DPR结果调用相应的模糊控制器进行混合电力电子系统的功率分配。与模糊逻辑控制策略相比，该方法的总成本降低了至少9.94%，而与动态规划策略相比，总成本的降低幅度在1.83%以内，表明该方法具有接近最优的实时性和可行性。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

IEEE Transactions on Power Electronics 工程技术-工程：电子与电气

CiteScore

15.20

自引率

20.90%

发文量

1099

审稿时长

3 months

期刊介绍： The IEEE Transactions on Power Electronics journal covers all issues of widespread or generic interest to engineers who work in the field of power electronics. The Journal editors will enforce standards and a review policy equivalent to the IEEE Transactions, and only papers of high technical quality will be accepted. Papers which treat new and novel device, circuit or system issues which are of generic interest to power electronics engineers are published. Papers which are not within the scope of this Journal will be forwarded to the appropriate IEEE Journal or Transactions editors. Examples of papers which would be more appropriately published in other Journals or Transactions include: 1) Papers describing semiconductor or electron device physics. These papers would be more appropriate for the IEEE Transactions on Electron Devices. 2) Papers describing applications in specific areas: e.g., industry, instrumentation, utility power systems, aerospace, industrial electronics, etc. These papers would be more appropriate for the Transactions of the Society which is concerned with these applications. 3) Papers describing magnetic materials and magnetic device physics. These papers would be more appropriate for the IEEE Transactions on Magnetics. 4) Papers on machine theory. These papers would be more appropriate for the IEEE Transactions on Power Systems. While original papers of significant technical content will comprise the major portion of the Journal, tutorial papers and papers of historical value are also reviewed for publication.