“大数据”时代的到来为风机健康监测带来了新机遇，然而，风机往往运行在极端恶劣环境下，监测数据中夹杂了大量缺失值，数据质量无法保障，进而会制定有误的运维指导策略。为保证风速监测数据质量，提出了基于高维时空张量CP分解的风速监测数据缺失值恢复方法。首先构建包含时空信息的四阶张量，接着，利用CP分解将张量分解为多个因子矩阵，再通过加权张量将恢复缺失数据转化为求解目标函数最小值，最后根据因子矩阵重构张量，从而获得缺失处原始信息值。利用提出方法与GPR、GRU、LSTM、SWLSTM等传统方法对某风电场的缺失数据进行恢复，结果表明：相比传统方法，提出方法的R2最接近1，MAE等误差指标均为最小，具有最高拟合度，从而验证了提出方法有效性。
Abstract
The advent of "big data" era brings new opportunities for wind turbine health condition monitoring, but wind turbines often operate in harsh environment, and thus monitoring data is mixed with a large number of missing values which reduces data quality. As a result, wrong operation and maintenance strategies will be developed based on these low-quality data. A method based on CP decomposition of spatial-temporal tensor is proposed to recover missing data to improve the quality of monitoring data. First, a four-dimension tensor containing spatial-temporal information is constructed. Then CP decomposition is applied to decompose the estimating tensor into factor matrices. Afterwards, a weighted tensor is used to translate the recovery issue into the solving of a minimization function. Finally, the tensor can be reconstructed according to the factor matrices, and the original value of the missing data can be obtained. The actual monitoring data of a wind farm is used to recover the missing values with different methods including GPR,GRU,LSTM,SWLSTM. Results show that the R2 of proposed method is closest to 1 and the other recovery error such as MAE are minimum, which has higher fitting degree with the real data. Therefore, the case verifies the effectiveness of the proposed method.
关键词
风机健康监测 /
数据质量 /
缺失值数据 /
张量分解 /
数据恢复
{{custom_keyword}} /
Key words
health condition monitoring of wind turbine /
data quality /
data with missing values /
tensor decomposition /
data recovery
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 雷亚国,贾峰,孔德同，等. 大数据下机械智能故障诊断的机遇与挑战[J]. 机械工程学报,2018, 54(05): 94-104.
LEI Yaguo, JIA Feng, KONG Detong, et al. Opportunities and challenges of machinery intelligent fault diagnosis in big data era[J]. Journal of Mechanical Engineering, 2018, 54(5): 94-104.
[2] YE Z, YU J B. AKSNet: A novel convolutional neural network with adaptive kernel width and sparse regularization for machinery fault diagnosis[J]. Journal of Manufacturing Systems, 2021, 59: 467-480.
[3] HAN T, LIU C, WU R. Deep transfer learning with limited data for machinery fault diagnosis[J]. Applied Soft Computing Journal,2021, 103: 107150.
[4] 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.
ZHAO Hongshan, LIU Huihai, LIU Hongyang, et al. Condition monitoring and fault diagnosis of wind turbine generator based on stacked autoencoder network[J]. Automation of Electric Power Systems, 2018, 42(11): 102-108.
[5] ANTONIO J Z, ALEJANDRO C, FRANCISCO G. Wind missing data arrangement using wavelet based techniques for getting maximum likelihood[J]. Energy Conversion and Management, 2019, 185: 552-561.
[6] FATIH O H, YUSUF O, MEHMET K. Missing Wind data forecasting with adaptive neuro-fuzzy inference system[J]. Neural Computing and Applications, 2009, 18(3): 207-212.
[7] 梅勇,李霄,胡在春,等. 基于风电机组控制原理的风功率数据识别与清洗方法[J]. 动力工程学报,2021, 41(04): 316-322.
MEI Yong, LI Xiao, HU Zaichun, et al. Identification and cleaning of wind power data methods based on control principle of wind turbine generator system[J]. Journal of Chinese Society of Power Engineering, 2021, 41(04): 316-322.
[8] 胡阳,乔依林. 基于置信等效边界模型的风功率数据清洗方法[J]. 电力系统自动化,2018, 42(15): 18-23.
HU Yang, QIAO Yilin. Wind power data cleaning.
Method based on confidence equivalent boundary model[J]. Automation of Electric Power Systems, 2018, 42(15): 18-23.
[9] XU X F, LEI Y G, LI Z D, An incorrect data detection method for big data cleaning of machinery condition monitoring[J]. IEEE Transactions on Industrial Electronics, 2020 ,67(3): 2326-2336.
[10] 陈帅, 赵明, 郭栋，等. 基于SVD-KDR算法的工业监测数据插补技术[J]. 机械工程学报, 2021, 57(2): 30-38.
CHEN Shuai, ZHAO Ming, GUO Dong, et al. Missing data imputation using SVD-KDR algorithm in industrial monitoring data[J]. Journal of Mechanical Engineering, 2021, 57(2): 30-38.
[11] YANG Z L, LIU Y Q, LI C R. Interpolation of missing wind data based on ANFIS[J]. Renewable Energy,2010, 36(3): 993-998.
[12] DU J, SUN H C, CAO Y J, et al. Ensemble interpolation of missing wind turbine nacelle wind speed data in wind farms based on robust particle swarm optimized generalized regression neural network[J]. International Journal of Green Energy,2019, 16(14): 1210-1219.
[13] 雷亚国,许学方,蔡潇,等. 面向机械装备健康监测的数据质量保障方法研究[J]. 机械工程学报,2021, 57(4): 1-9.
LEI Yaguo, XU Xuefang, CAI Xiao, et al. Research on data quality assurance for health condition monitoring of machinery[J]. Journal of Mechanical Engineering, 2021, 57(4): 1-9.
[14] 许小伟,沈琪,严运兵，等. 基于张量Tucker分解的发动机故障诊断[J]. 中国机械工程,2018, 29(05): 552-557.
XU Xiaowei, SHEN Qi, YAN Yunbing, et al. Engine fault diagnosis based on tensor Tucker decomposition[J]. China Mechanical Engineering, 2018, 29(05): 552-557.
[15] QIN L, LIU S, KANG Y, et al. Wind velocity distribution reconstruction using CFD database with Tucker decomposition and sensor measurement[J]. Energy,2019, 167: 1236-1250.
[16] TAN H C, FENG G D, FENG G S. A Tensor-based method for missing traffic data completion[J]. Transportation Research Part C,2013, 28: 15-27.
[17] 韩晓娟，曹慧，李勇，等. 基于小波变换和LS-SVM的短期风速预测方法[J]. 太阳能学报, 2011, 32(10): 1538-1542.
HAN Xiaojuan, CAO Hui, LI Yong, et al. Short term wind speed prediction based on wavelet transform and LS-SVM[J]. Acta Energiae Solaris Sinica, 2011, 32(10): 1538-1542.
[18] ACAR E, DUNLAVY D M, KOLDA T G, et al. Scalable tensor factorizations for incomplete data[J]. Chemometrics and Intelligent Laboratory Systems, 2011, 106(1): 41-5.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
PDF(2459 KB)
