基于数据融合和Wiener过程的风电轴承剩余寿命预测  

REMAINING LIFE PREDICTION OF WIND TURBINE BEARINGS BASED ON DATA FUSION AND WIENER PROCESSES

作　　者：杨志凌[1] 刘俊华 Yang Zhiling;Liu Junhua(School of Energy Power and Mechanical Engineering,North China Electric Power University,Beijing 102206,China)

机构地区：[1]华北电力大学能源动力与机械工程学院,北京102206

出　　处：《太阳能学报》2021年第10期189-194,共6页Acta Energiae Solaris Sinica

基　　金：国家重点研发计划政府间国际科技创新合作重点专项项目(2017YFE0109000)。

摘　　要：为实时掌握风电机组轴承剩余寿命,提出基于数据融合和Wiener过程的风电轴承剩余寿命预测方法。该方法采用主成分分析(PCA)方法融合多个特征参数进行Wiener过程建模;为减少对大量历史数据的依赖,使用Bootstrap抽样方法求取先验分布参数,贝叶斯(Bayes)方法在线更新模型参数。对比时域单特征量、时域多特征量及时域频域多特征量,发现基于多特征量的Wiener建模方法预测精度更高,该方法适用于新建风电场的风电机组轴承等关键部件的在线可靠性评估及剩余寿命预测。In order to obtain the remaining life of wind turbine bearings in real time,an assessment method based on data fusion and Wiener process is proposed.Firstly,the principal component analysis(PCA)method fuses multiple feature parameters for Wiener process modeling;Secondly,Bootstrap Sampling method is used to obtain prior distribution parameters for avoiding a large number of historical data of similar systems.Furthermore,Bayes method is used to update the model parameters online.Comparing both methods based on single feature and multi-feature,the Wiener process modeling method based multi-features data fusion improves the prediction accuracy.The proposed method is suitable for online reliability evaluation and remaining life assessment of key components such as wind turbine bearings in new wind farms.

关 键 词：风电机组 轴承 剩余寿命评估 WIENER过程 数据融合 

分 类 号：TK83[动力工程及工程热物理—流体机械及工程]