摘要
文章针对风电机组齿轮箱油温劣化特征识别问题,提出了一种基于工况细化的异常变化检测和故障早期预警方法。该方法根据风机叶轮转速将机组运行数据进行细化分仓,在每个叶轮转速仓中建立基于概率统计分析的齿轮箱油温正常行为模型并设定其温度分布和温升变化的异常阈值;然后对现场机组齿轮箱油温变化进行监测,利用时序滑动窗口的评估方式实现风机齿轮箱油温故障预警。
Aiming at the problem of lubricating oil temperature degradation characteristics identification of wind turbine gearbox,this paper proposes a method based on the refinement of wind turbine operational conditions about the abnormal change detection and the early warning of failure.The method realizes the operational conditions division according to the rotor speed under power production state of wind turbines,establishes a normal behavior model based on the statistics analysis of the gearbox oil temperature in each rotor speed bin and also sets the abnormal thresholds of the temperature distributions and temperature rise changes.Through the monitoring of the gearbox oil temperature changes of the on-site units,the time-sliding window evaluation method will be used to realize the early warning of gearbox oil temperature failure.
作者
董健
柳亦兵
滕伟
马志勇
Dong Jian;Liu Yibing;Teng Wei;Ma Zhiyong(Guodian United Power Technology Co.,Ltd.,Beijing 100039,China;Key Laboratory of Power Station Energy Transfer Conversion and System,North China Electric Power University,Beijing 102206,China)
出处
《可再生能源》
CAS
CSCD
北大核心
2021年第4期501-506,共6页
Renewable Energy Resources
基金
河北省科技计划项目(15214307D)。
关键词
风电机组
齿轮箱
故障预警
正常行为模型
工况细化
wind turbine
gearbox
fault early warning
normal behavior model
operational condition division