摘要
综述海上风力机响应极值预报的研究进展,重点介绍基于统计分析和机器学习的极值预报方法。统计方法包括渐进极值法、阈值超越法、平均穿越率法和平均条件超越率法。统计方法通过样本数据建模风力机响应的概率分布模型,进行极值预报。基于...展开更多
综述海上风力机响应极值预报的研究进展,重点介绍基于统计分析和机器学习的极值预报方法。统计方法包括渐进极值法、阈值超越法、平均穿越率法和平均条件超越率法。统计方法通过样本数据建模风力机响应的概率分布模型,进行极值预报。基于机器学习的极值预报方法(如人工神经网络和高斯过程回归等算法),通过提取数据特征并构建非线性模型,进行海上风力机响应极值预报,最后对各类极值预报方法的优缺点及适用场景进行总结,并展望极值预报技术的未来发展方向。收起
This paper reviews recent developments in short-term extreme value prediction for offshore wind turbine responses,with a particular focus on statistical and machine learning methods.Statistical methods,including the Gumbel distribution,Peak Over Threshold(POT),Mean Upcr...MORE
This paper reviews recent developments in short-term extreme value prediction for offshore wind turbine responses,with a particular focus on statistical and machine learning methods.Statistical methods,including the Gumbel distribution,Peak Over Threshold(POT),Mean Upcrossing,and Average Conditional Exceedance Rate(ACER)methods,model turbine response probability distributions based on sample data for short-term extreme value estimation.Additionally,machine learning techniques,such as artificial neural networks and Gaussian process regression,improve prediction accuracy by automatically extracting relevant data features and constructing nonlinear models.This review highlights the strengths and limitations of various short-term extreme value prediction methods,discusses their applicable scenarios,and outlines potential future directions for advancing prediction technologies.FEWER
作者
柴威
何林
施伟
陈威
曾佳焱
杨清泉
Chai Wei;He Lin;Shi Wei;Chen Wei;Zeng Jiayan;Yang Qingquan(Key Laboratory of High Performance Ship Technology,Ministry of Education(Wuhan University of Technology),Wuhan 430063,China;School of Naval Architecture,Ocean and Energy Power Engineering,Wuhan University of Technology,Wuhan 430063,China;Faculty of Infrastructure Engineering,Dalian University of Technology,Dalian 116024,China)
出处
《太阳能学报》
EI
CAS
CSCD
北大核心
2026年第3期308-315,共8页
Acta Energiae Solaris Sinica
基金
国家自然科学基金(52201379
52071058)
中央高校基本科研业务费专项资金(3120624109)。
关键词
海上风力机
响应
极值预报
统计方法
机器学习
代理模型
offshore wind turbines
responses
extreme value prediction
statistical methods
machine learning
surrogate model