摘要
针对风电机组工作环境恶略、故障频发的问题,提出了一种风电机组变桨系统的故障预警方法,将机器学习应用于故障智能诊断,使用支持向量回归算法建立了变桨系统的故障预警模型,同时,采用信息熵以及均方根误差作为预警的指标,可以更准确直观地对潜在故障做出预警。以新疆某风电厂运行数据进行仿真试验,仿真结果表明所建模型可在故障发生前8天对潜在故障做出预警,验证了预警模型的可行性和有效性。
Aiming at the problem of bad working environment and frequent faults of wind turbines,a fault early warning method for pitch control system of wind turbines is proposed. Machine learning is applied to fault intelligent diagnosis. A fault early warning model of pitch control system is established by using support vector regression algorithm. Meanwhile,information entropy and root mean square error are used as the indicators,which can be more accurate and intuitive to make early warning of potential failures. The simulation results of a wind power plant in Xinjiang show that the model can warn potential faults 8 days in advance,which verifies the feasibility and validity of the early warning model.
作者
王伟
吕丽霞
张厚
WANG Wei;LU Lixia;ZHANG Hou(School of Control and Computer Engineering,North China Electric Power University,Baoding 071003,China)
出处
《电力科学与工程》
2019年第10期73-78,共6页
Electric Power Science and Engineering