网络出版:2025-10-31,
纸质出版:2025-10-31
移动端阅览
许伯强, 王彪, 孙丽玲, 等. 基于并行架构网络与改进动态FD-KNN的风力发电机轴承故障预警[J]. 太阳能学报, 2025,(10):753-765.
许伯强, 王彪, 孙丽玲, et al. 基于并行架构网络与改进动态FD-KNN的风力发电机轴承故障预警[J]. 2025, (10): 753-765.
许伯强, 王彪, 孙丽玲, 等. 基于并行架构网络与改进动态FD-KNN的风力发电机轴承故障预警[J]. 太阳能学报, 2025,(10):753-765. DOI: doi:10.19912/j.0254-0096.tynxb.2024-1121.
许伯强, 王彪, 孙丽玲, et al. 基于并行架构网络与改进动态FD-KNN的风力发电机轴承故障预警[J]. 2025, (10): 753-765. DOI: doi:10.19912/j.0254-0096.tynxb.2024-1121.
针对当前风力发电机轴承故障预警准确率和可靠性不足的问题
提出一种基于并行架构网络与改进动态k近邻故障检测(FD-KNN)的风力发电机轴承故障预警方法。首先
对风力发电机的数据采集与监控(SCADA)数据进行相关性分析
筛选出与风力发电机轴承关键变量高度相关的变量
并采用集合模态经验分解(EEMD)分解关键变量
深入挖掘关键变量内不同时间尺度的特征以及关键变量与高相关协变量的潜在相互作用。然后
构建一个结合自注意力机制的长短期记忆网络(Self-Attention-LSTM)和改进Transformer模型的新型并行架构网络
用以精确可靠地预测关键变量的未来状态。基于预测结果
计算残差
并结合风力发电机轴承的实时状态对FD-KNN算法进行动态优化
包括调整近邻规模、设置动态告警阈值和预警条件
以实现更为精准可靠的故障预警。最后
通过实际SCADA数据验证
结果表明该方法可提前识别风力发电机轴承故障
且在准确性和可靠性方面均表现出色。
Aiming at the current problem of insufficient accuracy and reliability of wind turbine bearing fault warning
a wind turbine bearing fault warning method based on parallel architecture network with improved dynamic fault detection--nearest neighbor (FD-KNN) is proposed. Firstly
the correlation analysis of supervisory control and data acquisition (SCADA) of wind turbine is carried out to screen out the variables that are highly correlated with the key variables of wind turbine bearings
and the key variables are decomposed by using the Ensemble empirical mode decomposition (EEMD)
to dig deeply the characteristics of the key variables at different time scales and the potential interactions between key variables and highly correlated covariates. Then
a novel parallel architecture network that combines Self-Attention LSTM and an improved Transformer model is developed to accurately and reliably predict the future states of key variables. Based on the prediction results
the residuals are calculated
and the FD-KNN algorithm is dynamically optimized using the real-time state of wind turbine bearings. This optimization includes adjusting the scale of the nearest neighbors and setting dynamic warning thresholds and conditions to achieve more accurate and reliable fault warnings. Finally
the method is validated using actual SCADA data
and the results demonstrate that it can identify wind turbine bearing faults in advance
excelling in both accuracy and reliability.
Global Wind Energy Council. Global wind report 2023[R]. Brussels: GWEC-Global Wind Energy Council, 2024.
FU J, CHU J C, GUO P, et al.Condition monitoring of wind turbine gearbox bearing based on deep learning model[J]. IEEE access, 2019, 7: 57078-57087.
KANDUKURI S T, SENANYAKA J S L, HUYNH V K, et al. A two-stage fault detection and classification scheme for electrical pitch drives in offshore wind farms using support vector machine[J]. IEEE transactions on industry applications, 2019, 55(5): 5109-5118.
沈小军, 付雪姣, 周冲成, 等. 风电机组风速-功率异常运行数据特征及清洗方法[J]. 电工技术学报, 2018, 33(14): 3353-3361.
CROSS P, MA X D.Nonlinear system identification for model-based condition monitoring of wind turbines[J]. Renewable energy, 2014, 71: 166-175.
YAMPIKULSAKUL N, BYON E, HUANG S, et al.Condition monitoring of wind power system with nonparametric regression analysis[J]. IEEE transactions on energy conversion, 2014, 29(2): 288-299.
SUN P, LI J, WANG C S, et al.A generalized model for wind turbine anomaly identification based on SCADA data[J]. Applied energy, 2016, 168: 550-567.
SKRIMPAS G A, SWEENEY C W, MARHADI K S, et al.Employment of kernel methods on wind turbine power performance assessment[J]. IEEE transactions on sustainable energy, 2015, 6(3): 698-706.
向玲, 王朋鹤, 李京蓄. 基于CNN-LSTM的风电机组异常状态检测[J]. 振动与冲击, 2021, 40(22): 11-17.
王依宁, 解大, 王西田, 等. 基于PCA-LSTM模型的风电机网相互作用预测[J]. 中国电机工程学报, 2019, 39(14): 4070-4081.
SUN Z X, SUN H X.Stacked denoising autoencoder with density-grid based clustering method for detecting outlier of wind turbine components[J]. IEEE access, 2019, 7: 13078-13091.
刘家瑞, 杨国田, 杨锡运. 基于深度卷积自编码器的风电机组故障预警方法研究[J]. 太阳能学报, 2022, 43(11): 215-223.
陈俊生, 李剑, 陈伟根, 等. 采用滑动窗口及多重加噪比堆栈降噪自编码的风电机组状态异常检测方法[J]. 电工技术学报, 2020, 35(2): 346-358.
李东东, 刘宇航, 赵阳, 等. 基于改进生成对抗网络的风机行星齿轮箱故障诊断方法[J]. 中国电机工程学报, 2021, 41(21): 7496-7507.
王桂兰, 赵洪山, 米增强. XGBoost算法在风机主轴承故障预测中的应用[J]. 电力自动化设备, 2019, 39(1): 73-77, 83.
马良玉, 程善珍. 基于支持向量数据描述和XGBoost的风电机组异常工况预警研究[J]. 电工技术学报, 2022, 37(13): 3241-3249.
赵洪山, 刘辉海. 基于性能改善深度置信网络的风电机组主轴承状态分析[J]. 电力自动化设备, 2018, 38(2): 44-49.
尹诗, 侯国莲, 于晓东, 等. 基于SCADA数据的风电机组齿轮箱状态监测方法[J]. 太阳能学报, 2021, 42(1): 324-332.
陈尚年, 李录平, 张世海, 等. 基于EEMD-LSTM的汽轮机转子碰磨故障诊断模型及其工程应用[J]. 热能动力工程, 2023, 38(8): 159-168.
赵洪山, 林诗雨, 孙承妍, 等. 考虑多时间尺度信息的风力发电机滚动轴承故障预测[J]. 中国电机工程学报, 2024, 44(22): 8908-8920.
XIANG L, WANG P H, YANG X, et al.Fault detection of wind turbine based on SCADA data analysis using CNN and LSTM with attention mechanism[J]. Measurement, 2021, 175: 109094.
胡爱军, 连俭, 向玲. 基于ACNN和Bi-LSTM的风电机组故障早期识别[J]. 太阳能学报, 2021, 42(12): 143-149.
俞国燕, 李少伟, 董晔弘. 基于XGBoost-LightGBM-LSTM的风机齿轮箱轴承故障预警[J]. 轴承, 2023(6): 140-145.
庞昊, 高金峰, 杜耀恒. 基于多神经网络融合的短期负荷预测方法[J]. 电力自动化设备, 2020, 40(6): 37-43.
张书瑶, 王梓齐, 刘长良. 基于改进集成KNN回归算法的风电机组齿轮箱状态监测[J]. 动力工程学报, 2023, 43(6): 759-767.
LIM B, ARıK S Ö, LOEFF N, et al. Temporal fusion Transformers for interpretable multi-horizon time series forecasting[J]. International journal of forecasting, 2021, 37(4): 1748-1764.
VASWANI A, SHAZEER N, PARMAR N, et al.Attention is all you need[J]. Advances in neural information processing systems, 2017, 30: 5998-6008.
钱小毅, 孙天贺, 姜兴宇, 等. 基于动态邻域指标重构的风电机组故障检测方法[J]. 太阳能学报, 2024, 45(9): 557-563.
0
浏览量
0
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621