吉庆昌 邸英杰 阴兆武 张冬梅



【摘 要】针对风电机组运行中状态监测的特点和开展计划性维护与维修的需求,论文应用大数据处理技术设计了一种风电机组故障诊断及预警系统。系统集成了基于风电机组海量运行状态监测数据的数据采集、数据挖掘、机器学习等技术,由数据采集与汇聚、数据挖掘与计算、状态展示与应用3个子系统组成,可以自动地计算出反映机组运行状态的特征值,将不同的特征值进行逻辑组合可实现不同故障的诊断与预警,以便对风电机组开展计划性维护与维修。
【Abstract】According to the characteristics of condition monitoring in wind turbine operation and the needs to carry out planned maintenance and repair, the paper applies big data processing technology to design a fault diagnosis and early warning system for wind turbine. The system integrates data acquisition, data mining, machine learning and other technologies based on the massive operation condition monitoring data of wind turbine. It is composed of three subsystems: data acquisition and aggregation, data mining and calculation, condition display and application, which can automatically calculate the eigenvalues reflecting the operation condition of the turbine. The logical combination of different eigenvalues can realize different fault diagnosis and early warning, so as to carry out planned maintenance and repair of wind turbine.
【关键词】风电机组;故障预警;大数据处理技术
【Keywords】wind turbine; fault early warning; big data processing technology
【中图分类号】TM315 【文献标志码】A 【文章编号】1673-1069(2021)10-0179-03
1 引言
风能作为一种清洁能源,相对于其他可再生能源,在经济效益、技术成熟度、基础设施建设等方面具有明显优势,它是全球发展较为迅速的新能源发电技术[1]。面对日趋严峻的碳减排形势,为早日实现碳达峰和碳中和目标,我国风电装机规模继续保持稳步增长。由于对风能的特殊需求,我国的风电场多分布在哈密、蒙东和蒙西、河北坝上、东南沿海岛屿等偏远地区。另外,随着风电机组使用年限的增长,各种故障频发,于是对风电机组的运行状态进行在线监测,实现故障诊断与预警,开展计划性维护与维修显得越来越重要。
风电机组在线监测与故障诊断技术对于及时了解机组的运行状态、对机组故障进行预警、有针对性地开展设备维护、减少突发性故障和提高设备利用率都具有重要的现实意义。……