Quantum machine learning based wind turbine condition monitoring: State of the art and future prospects

IF 9.9 1区 工程技术 Q1 ENERGY & FUELS
Zhefeng Zhang, Yueqi Wu, Xiandong Ma
{"title":"Quantum machine learning based wind turbine condition monitoring: State of the art and future prospects","authors":"Zhefeng Zhang,&nbsp;Yueqi Wu,&nbsp;Xiandong Ma","doi":"10.1016/j.enconman.2025.119694","DOIUrl":null,"url":null,"abstract":"<div><div>Wind energy, as a popular renewable resource, has gained extensive development and application in recent decades. Effective condition monitoring and fault diagnosis are crucial for ensuring the reliable operation of wind turbines. While conventional machine learning methods have been widely used in wind turbine condition monitoring, these approaches often face challenges such as complex feature extraction, limited model generalization, and high computational costs when dealing with large-scale, high-dimensional, and complex datasets. The emergence of quantum computing has opened up a new paradigm of machine learning algorithms. Quantum machine learning combines the advantages of quantum computing and machine learning, with the potential to surpass classical computational capabilities. This paper firstly reviews applications and limitations of the state-of-the-art machine learning-based condition monitoring techniques for wind turbines. It then reviews the fundamentals of quantum computing, quantum machine learning algorithms and their applications, covering quantum-based feature extraction, classification and regression for fault detection and the use of quantum neural networks for predictive maintenance. Through comparison, it is observed that quantum machine learning methods, even without extensive optimization, can achieve accuracy levels comparable to those of optimized conventional machine learning approaches. The challenges of applying quantum machine learning are also addressed, along with the future research and development prospects. The objective of this review is to fill a gap in the published literature by providing a new paradigm approach for wind turbine condition monitoring. By promoting quantum machine learning in this field, the reliability and efficiency of wind power systems are ultimately sought to be enhanced.</div></div>","PeriodicalId":11664,"journal":{"name":"Energy Conversion and Management","volume":"332 ","pages":"Article 119694"},"PeriodicalIF":9.9000,"publicationDate":"2025-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Energy Conversion and Management","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0196890425002171","RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENERGY & FUELS","Score":null,"Total":0}
引用次数: 0

Abstract

Wind energy, as a popular renewable resource, has gained extensive development and application in recent decades. Effective condition monitoring and fault diagnosis are crucial for ensuring the reliable operation of wind turbines. While conventional machine learning methods have been widely used in wind turbine condition monitoring, these approaches often face challenges such as complex feature extraction, limited model generalization, and high computational costs when dealing with large-scale, high-dimensional, and complex datasets. The emergence of quantum computing has opened up a new paradigm of machine learning algorithms. Quantum machine learning combines the advantages of quantum computing and machine learning, with the potential to surpass classical computational capabilities. This paper firstly reviews applications and limitations of the state-of-the-art machine learning-based condition monitoring techniques for wind turbines. It then reviews the fundamentals of quantum computing, quantum machine learning algorithms and their applications, covering quantum-based feature extraction, classification and regression for fault detection and the use of quantum neural networks for predictive maintenance. Through comparison, it is observed that quantum machine learning methods, even without extensive optimization, can achieve accuracy levels comparable to those of optimized conventional machine learning approaches. The challenges of applying quantum machine learning are also addressed, along with the future research and development prospects. The objective of this review is to fill a gap in the published literature by providing a new paradigm approach for wind turbine condition monitoring. By promoting quantum machine learning in this field, the reliability and efficiency of wind power systems are ultimately sought to be enhanced.

Abstract Image

基于量子机器学习的风力涡轮机状态监测:技术现状与未来展望
风能作为一种受欢迎的可再生能源,近几十年来得到了广泛的开发和应用。有效的状态监测和故障诊断是保证风力发电机组可靠运行的关键。虽然传统的机器学习方法已广泛应用于风力涡轮机状态监测,但这些方法在处理大规模、高维和复杂的数据集时往往面临复杂的特征提取、有限的模型泛化和高计算成本等挑战。量子计算的出现开启了机器学习算法的新范式。量子机器学习结合了量子计算和机器学习的优势,具有超越经典计算能力的潜力。本文首先回顾了基于机器学习的风力涡轮机状态监测技术的应用和局限性。然后回顾量子计算,量子机器学习算法及其应用的基础知识,涵盖基于量子的特征提取,故障检测的分类和回归以及用于预测性维护的量子神经网络的使用。通过比较,我们观察到量子机器学习方法,即使没有广泛的优化,也可以达到与优化后的传统机器学习方法相当的精度水平。讨论了应用量子机器学习的挑战,以及未来的研究和发展前景。这篇综述的目的是通过为风力涡轮机状态监测提供一种新的范式方法来填补已发表文献的空白。通过在该领域推广量子机器学习,最终寻求提高风力发电系统的可靠性和效率。
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
Energy Conversion and Management
Energy Conversion and Management 工程技术-力学
CiteScore
19.00
自引率
11.50%
发文量
1304
审稿时长
17 days
期刊介绍: The journal Energy Conversion and Management provides a forum for publishing original contributions and comprehensive technical review articles of interdisciplinary and original research on all important energy topics. The topics considered include energy generation, utilization, conversion, storage, transmission, conservation, management and sustainability. These topics typically involve various types of energy such as mechanical, thermal, nuclear, chemical, electromagnetic, magnetic and electric. These energy types cover all known energy resources, including renewable resources (e.g., solar, bio, hydro, wind, geothermal and ocean energy), fossil fuels and nuclear resources.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信