中国科技核心期刊
中国建筑科学领域高质量科技期刊分级目录T2级期刊
RCCSE中国核心学术期刊
美国化学文摘社（CAS）数据库 收录期刊
日本JST China 收录期刊
世界期刊影响力指数（WJCI）报告 收录期刊

留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

基于深度学习的海上风机多源缺失数据的填补方法

李栋,  廖毅桢,  桑源,  李泽宇,  姚博,  陈洪兵

downloadPDF
文章导航 > 工业建筑  > 2026 >  56(5): 176-186
李栋, 廖毅桢, 桑源, 李泽宇, 姚博, 陈洪兵. 基于深度学习的海上风机多源缺失数据的填补方法[J]. 工业建筑, 2026, 56(5): 176-186. doi: 10.3724/j.gyjzG26031308
引用本文: 李栋, 廖毅桢, 桑源, 李泽宇, 姚博, 陈洪兵. 基于深度学习的海上风机多源缺失数据的填补方法[J]. 工业建筑, 2026, 56(5): 176-186. doi: 10.3724/j.gyjzG26031308
shu
LI Dong, LIAO Yizhen, SANG Yuan, LI Zeyu, YAO Bo, CHEN Hongbing. Deep Learning-Based Imputation of Multi-Source Missing Data for Offshore Wind Turbines[J]. INDUSTRIAL CONSTRUCTION, 2026, 56(5): 176-186. doi: 10.3724/j.gyjzG26031308
Citation: LI Dong, LIAO Yizhen, SANG Yuan, LI Zeyu, YAO Bo, CHEN Hongbing. Deep Learning-Based Imputation of Multi-Source Missing Data for Offshore Wind Turbines[J]. INDUSTRIAL CONSTRUCTION, 2026, 56(5): 176-186. doi: 10.3724/j.gyjzG26031308
shu

基于深度学习的海上风机多源缺失数据的填补方法

doi: 10.3724/j.gyjzG26031308

  • 1. 福州大学土木工程学院, 福州 350108;

  • 2. 北京科技大学未来城市学院/城镇化与城市安全研究院, 北京 100083;

  • 3. 中国五冶集团有限公司, 成都 476300

详细信息
    作者简介:

    李栋,副研究员,主要从事大跨度空间薄膜结构、组合结构、海上风电、结构风工程等方面研究,dongli@fzu.edu.cn。

    通讯作者:

    陈洪兵,教授,主要从事工程结构损伤检测、服役安全智能诊断,钢-混凝土组合结构缺陷无损检测技术和多尺度多物理场仿真等领域研究,hongbingchen@ustb.edu.cn。

Deep Learning-Based Imputation of Multi-Source Missing Data for Offshore Wind Turbines

  • 1. College of Civil Engineering, Fuzhou University, Fuzhou 350108, China;

  • 2. Future City College/Institute of Urbanization and Urban Safety, University of Science and Technology Beijing, Beijing 100083, China;

  • 3. China MCC5 Group Co., Ltd., Chengdu 476300, China

    摘要
  • 摘要: 针对海上风电机组在恶劣工况下因传感器失效或通信中断引发的多源监测数据缺失问题，构建了一种基于多头门控残差网络的填补模型。该方法通过特征级联将数据采集与监视控制系统（SCADA）数据和结构振动监测数据进行协同融合，并引入门控残差网络提取深度非线性耦合特征；同时设计出多头并行输出结构，实现对两类异构数据的独立重建。训练阶段采用动态掩码机制与混合损失函数，以增强模型对复杂气动工况的适应能力。基于某10 MW海上风电机组实测数据的验证结果显示，该模型在训练工况下具有较高的决定系数，能够实现多源缺失数据的精准重构。在面向非训练时段的泛化测试中，模型预测的决定系数虽略有波动，但仍能有效捕捉信号整体变化趋势；尤其对振动类数据，其泛化性能衰减幅度小于SCADA数据，展现出更强的稳定性。该方法可显著提升风机多源监测数据的完整性与可靠性，具备良好的工程应用前景。
    Abstract: To address the problem of missing multi-source monitoring data of offshore wind turbines caused by sensor failures or communication interruptions under harsh operating conditions， this paper proposes a novel imputation model based on a multi-head gated residual network. This method achieves collaborative fusion of supervisory control and data acquisition （SCADA） data and structural vibration monitoring data through feature concatenation， employs a gated residual network to extract deep nonlinear coupling features， and uses a multi-head parallel output architecture for the independent reconstruction of these two heterogeneous data types. During the training stage， a dynamic masking mechanism combined with a hybrid loss function is adopted to enhance the model’s adaptability to complex aerodynamic operating conditions. Validated with field data from a 10 MW offshore wind turbine， the proposed model achieved a high coefficient of determination under training conditions， enabling accurate reconstruction of missing multi-source data. In generalization tests for non-training periods， although the coefficient of determination of the model’s predictions fluctuated slightly， the model still effectively captured the overall trends of monitoring signals. Notably， the degradation in generalization performance for vibration data was less pronounced than that for SCADA data， demonstrating its greater stability. The proposed method can significantly improve the completeness and reliability of multi-source monitoring data for wind turbines and holds considerable potential for engineering applications.
    • HTML全文
    • 参考文献(22)
  • [1] WANG F X，ZHANG S，HOU Y Z，et al. Extraction of offshore wind turbines in china by combining multispectral and SAR image data［J］. Journal of Selected Topics in Applied Earth Observations and Remote Sensing，2024，17：3392695.
    [2] 蔡彦枫，李晓宇，任宗栋，等. 面向空中风力发电系统的风能资源分析［J］. 太阳能学报，2025，46（2）：624- 632.
    [3] 张珺辉，董绍华，叶迎春. 基于LabVIEW的海上风机齿轮箱故障诊断系统设计［J］. 新技术新工艺，2025（1）：29- 37.
    [4] LEI Y G，YANG B，JIANG X W，et al. Applications of machine learning to machine fault diagnosis:a review and roadmap［J］. Mechanical Systems and Signal Processing，2020，138：106587.
    [5] TAO S Y，QIAN Z，PEI Y，et al. Wind turbine failure detection based on SCADA data and data mining method［C］// Renewable Power Generation Conference. 2019.
    [6] ABORGEL T，SHEHATA A S，KOTB M A，et al. Heavy lift semi-submersible ships utilization in offshore wind turbines industry［J］. Energy Reports，2022，8：834- 847.
    [7] ZHANG W W，VATN J，RASHEED A. Gearbox pump failure prognostics in offshore wind turbine by an integrated data-driven model［J］. Applied Energy，2025，380：124829.
    [8] HUSSAIN M，MIRJAT N H，SHAIKH F，et al. Condition monitoring and fault diagnosis of wind turbine：a systematic literature review［J］. Journal of Selected Topics in Applied Earth Observations and Remote Sensing，2024，35：14747.
    [9] GOROSTIDI N，PARDO D，NAVA V. Diagnosis of the health status of mooring systems for floating offshore wind turbines［J］. Renewable Energy and Power Quality Journal，2023，287：115862.
    [10] ENCALADA-DAVILA Á，TUTIVÉN C，MOYÓN L，et al. Main bearing fault prognosis in wind turbines based on gated recurrent unit neural networks［J］. Renewable Energy and Power Quality Journal，2022，20：2172- 2183.
    [11] ZHAO J H，SHEN X D，LIU T J，et al. Probabilistic imputation of missing offshore wind speed based on conditional diffusion models［J］. Sustainable Energy，Grids and Networks，2025，44：101949.
    [12] QU F M，LIU J H，MA Y J，et al. A novel wind turbine data imputation method with multiple optimizations based on GANs［J］. Mechanical Systems and Signal Processing，2020，139：106610.
    [13] LIAO W L，BAK-JENSEN B，PILLAI J R，et al. Data-driven missing data imputation for wind farms using context encoder［J］. Journal of Modern Power Systems and Clean Energy，2022（2）：000894.
    [14] FAN Y W，FENG C L，WU R，et al. Multiscale-attention masked autoencoder for missing data imputation of wind turbines［J］. Knowledge-Based Systems，2024，299：112114.
    [15] RADEV B，KASI V R，BINGI K. Enhancement of Texas wind turbine power predictions using fractional order neural network by incorporating machine learning models to impute missing data［J］. Knowledge-Based Systems，2024，300：112176.
    [16] WANG B，ZHOU B，ZOU M H，et al. Research on condition monitoring of offshore wind turbine gearbox based on K-means clustering and extreme learning machine［C］// International Conference on Artificial Intelligence and Power Engineering. 2023.
    [17] JU S H，HUANG Y C. Analyses of offshore wind turbine structures with soil-structure interaction under earthquakes［J］. Ocean Engineering，2019，187：106190.
    [18] GAO B，LI C，ZHOU F L，et al. Shaking table tests of offshore wind turbine systems with a suction bucket foundation in sandy seabed subject to earthquake and wind loads［J］. Marine Structures，2025，99：103706.
    [19] BAI H，ZHANG J H，SUN K，et al. Nonlinear long-term vibration response prediction of offshore wind turbines under full operating conditions based on deep learning［J］. Applied Ocean Research，2025，159：104625.
    [20] 赵洪山，刘辉海，刘宏杨，等. 基于堆叠自编码网络的风电机组发电机状态监测与故障诊断［J］. 电力系统自动化，2018，42（11）：102- 108.
    [21] 张舒翔，刘晓彤，郭旭峰，等. 基于SCADA数据驱动的风电机组状态监测［J］. 电工技术，2022（22）：47- 50.
    [22] *福建省自然科学基金面上项目（2024 J 01266）。
    • 相关文章
    • 施引文献
  • 资源附件(0)
  • 加载中
WeChat 点击查看大图
计量
  • 文章访问数:  8
  • HTML全文浏览量:  2
  • PDF下载量:  0
  • 被引次数: 0
出版历程
  • 收稿日期:  2026-03-13
  • 网络出版日期:  2026-06-06

目录

    /

    返回文章
    返回

    期刊在线

    作者中心

    友情链接

    版权所有 ©《工业建筑》杂志社有限公司 京ICP备11033253号-1

    本系统由北京仁和汇智信息技术有限公司 设计开发   百度统计