基于SCADA的风力机叶片PHM系统设计与实现

Design and Implementation of PHM System for Wind Turbine Blade Based on SCADA

随着我国风电产业的迅速发展,风机装机容量逐渐增加,叶片的长度也随之增长,所以风机产生故障的机率越来越大,因此,开展风力机叶片预测与健康管理(Prognostics and Health Management,简称PHM)的研究具有重要意义。首先,从监控与数据采集系统(Supervisory Control and Data Aequisition,简称SCADA)中提取叶片正常和异常状态下的特征参数,其次,采用BP神经网络建立叶片健康状态预测模型,最后采用Power Designer建模软件建立系统物理数据模型,并对系统架构进行设计,将训练好的神经网络模型应用在风力机叶片PHM系统中,有效提高了风力机叶片预测与健康管理的信息化水平。

With the rapid development of the wind power industry.the wind turbine installation capacity and the length of the blade are increased,resulting in inereased failure rate of the fan。Wind turbine blade prognostics and health management(PHM)is crucial in reducing the failure rate.First,the superv isory control and data acquisition(SCADA)is deployed to determine the characerstics in the normal and abnormal state.Secondly,the BP neural network is used to establish the blade health status prediction model.Finally,Power Designer software is used to establish a systematie data model and design the system architecture.By applying the trained neural network model in the wind turbine PHM system,more information can be extracted in wind turbine blade prediction and health management。