作者：刘帅^1,2, 刘长良², 曾华清³

作者单位：1. 华北电力大学控制与计算机工程学院, 北京 102206;
2. 华北电力大学 新能源电力系统国家重点实验室, 北京 102206;
3. 中国舰船研究设计中心, 湖北 武汉 430064

关键词：风电机组;故障预警;保局投影;核极限学习机;信息熵

摘要：

风电机组运行环境恶劣、机组设备衰退是近年来齿轮箱故障频发的主要原因，其设备状态与机组安全性、运营成本息息相关。面对这一挑战，利用监控与数据采集系统数据，提出一种将保局投影、核极限学习机和信息熵相结合的风电机组齿轮箱故障预警方法。采用保局投影对风电机组状态参数进行特征提取后，使用核极限学习机建立状态参数预测模型，最后辅以改进的加入信息熵概念，可准确预警异常工况。以河北省张家口某一风电场的运行数据作为实例进行研究，仿真结果表明，所提算法至少能提前2天预警潜在故障，验证该预警方法的有效性与实效性。

Research on fault warning for wind turbine gearbox based on kernel extreme learning machine
LIU Shuai^1,2, LIU Changliang², ZENG Huaqing³
1. School of Control and Computer Engineering, North China Electric Power University, Beijing 102206, China;
2. State Key Laboratory of Alternate Electrical Power System with Renewable Energy Sources, North China Electric Power University,Beijing 102206, China;
3. China Ship Development and Design Center, Wuhan 430064, China
Abstract: The harsh operating environment of wind turbines and the equipment degradation are the main reasons for the frequent failure of gearboxes in recent years. The equipment status is closely related to wind turbine safety and operating costs. In order to tackle the challenge, a fault warning method for wind turbine gearbox failure is proposed, which combines the locality preserving projections, the kernel extreme learning machine, and the information entropy. After that the feature extraction of the wind turbine's state parameters are carried out by using the locality preserving projections, the kernel extreme learning machine is applied to establish the state parameter prediction model. Finally, the improved information entropy concept is used, aiming at accurately predicting the abnormal working conditions. The operation data of a wind farm in Zhangjiakou, Hebei Province is studied as an example. The simulation results show that the proposed algorithm can warn potential faults at least 2 days in advance. The case study verifies the effectiveness and timeliness of the proposed fault warning method.
Keywords: wind turbines;fault warning;locality preserving projections;kernel extreme learning machine;information entropy
2019, 45(2):121-127  收稿日期: 2018-08-09;收到修改稿日期: 2018-09-21
基金项目: 北京市自然科学基金项目（4182061）；中央高校基本科研业专项资金（9163116001，2016MS143，2018ZD05）
作者简介: 刘帅(1990-),男,河北安国市人,博士研究生,研究方向为风电机组故障预警
参考文献
[1] DAWOUD S M, LIN X, OKBA M I. Hybrid renewable microgrid optimization techniques:A review[J]. Renewable and Sustainable Energy Reviews, 2017, 82(3):2039-2052
[2] ALHMOUD L, WANG B. A review of the state-of-the-art in wind-energy reliability analysis[J]. Renewable and Sustainable Energy Reviews, 2017, 81(2):1643-1651
[3] HAMEED Z, HONG Y, CHO Y, et al. Condition monitoring and fault detection of wind turbines and related algorithms:A review[J]. Renewable and Sustainable Energy Reviews, 2010, 13(1):1-39
[4] BANGALORE P, PATRIKSSON M. Analysis of SCADA data for early fault detection, with application to the maintenance management of wind turbines[J]. Renewable Energy, 2018, 115(1):521-532
[5] MARVUGLIA A, MESSINEO A. Monitoring of wind farms power curves using machine learning techniques[J]. Applied Energy, 2012, 98(10):574-83
[6] GRAY C S, WATSON S J. Physics of failure approach to wind turbine condition based maintenance[J]. Wind Energy, 2010, 13(5):395-405
[7] LAPIRA E, BRISSET D, ARDAKANI HD, et al. Wind turbine performance assessment using multi-regime modeling approach[J]. Renew Energy, 2012, 45(9):86-95
[8] QIU Y, FENG Y, SUN J, et al. Applying thermophysics for wind turbine drivetrain fault diagnosis using SCADA data[J]. IET Renewable Power Generation, 2016, 10(5):661-668
[9] DAO P B, STASZEWSKI W J, BARSZCZ T, et al. Condition monitoring and fault detection in wind turbines based on cointegration analysis of SCADA data[J]. Renewable Energy, 2018, 116(2):107-122
[10] BORCHERSEN A B, KINNAERT M. Model-based fault detection for generator cooling system in wind turbines using SCADA data[J]. Wind Energy, 2016, 19(4):593-606
[11] 曹梦楠, 邱颖宁, 冯延晖, 等. 基于无迹卡尔曼方法的风电机组齿轮箱故障诊断[J]. 太阳能学报, 2017, 38(1):32-38
[12] 张永辉. 基于监测数据的风力发电机故障预警研究[D]. 沈阳:沈阳工程学院, 2017.
[13] 董玉亮, 顾煜炯. 基于保局投影与自组织映射的风电机组故障预警方法[J]. 太阳能学报, 2015, 36(5):1123-1129
[14] 颜永龙, 李剑, 李辉, 等. 采用信息熵和组合模型的风电机组异常检测方法[J]. 电网技术, 2015, 39(3):737-743
[15] 刘帅, 刘长良, 甄成刚, 等. 基于群体多维相似性的风机齿轮箱预警策略[J]. 仪器仪表学报, 2018, 39(1):180-189
[16] HE X F, NIYOGI P. Locality preserving projections[A]. In:Proceeding of Neural Information Processing System[C], Vancouver, Canada, 2003:153-160.
[17] DING X, HE Q, LUO N. A fusion feature and its improvement based on locality preserving projections for rolling element bearing fault classification[J]. Journal of Sound and Vibration, 2015, 335(20):367-383
[18] SOLDERA J, BEHAINE C A R, SCHARCANSKI J. Customized orthogonal locality preserving projections with soft-margin maximization for face recognition[J]. IEEE Transactions on Instrumentation and Measurement, 2015, 64(9):2417-2426
[19] HE F, XU J. A novel process monitoring and fault detection approach based on statistics locality preserving projections[J]. Journal of Process Control, 2016, 37(1):46-57
[20] HUANG G B, ZHU Q Y, SIEW C K. Extreme learning machine:theory and applications[J]. Neurocomputing, 2006, 70(1/3):489-501
[21] ZENG Y, XU X, SHEN D, et al. Traffic sign recognition using kernel extreme learning machines with deep perceptual features[J]. IEEE Transactions on Intelligent Transportation Systems, 2017, 18(6):1647-1653
[22] BOUZGOU H, GUEYMARD C A. Minimum redundancy-Maximum relevance with extreme learning machines for global solar radiation forecasting:Toward an optimized dimensionality reduction for solar time series[J]. Solar Energy, 2017, 158:595-609
[23] MOUATADID S, ADAMOWSKI J. Using extreme learning machines for short-term urban water demand forecasting[J]. Urban Water Journal, 2017, 14(6):630-638
[24] LIU X L, XU X L, JIANG Z L, et al. Application of the state deterioration evolution based on bi-spectrum entropy and HMM in wind turbine[J]. Chaos, Solitons & Fractals, 2016, 89(8):160-168