基于改进融合的大型风力发电机组运行状态检测研究  

Research on the detection of the operating state of large-scale wind turbines based on improved fusion

作　　者：邓森 胡从星 梁太阳 杨朋雨 马先松 DENG Sen;HU Congxing;LIANG Taiyang;YANG Pengyu;MA Xiansong(State Power Investment Corporation Jiangsu Electric Power Co.,Ltd.,Nanjing 210000,China)

机构地区：[1]国家电投集团江苏电力有限公司,江苏南京210000

出　　处：《粘接》2024年第3期169-172,共4页Adhesion

基　　金：浙江省科技厅“尖兵”“领雁”研发攻关计划项目(项目编号:2022C01SA371625)。

摘　　要：为解决传统的大型风力发电机设备运行状态检测方法效果不佳的问题,设计了基于多特征融合的大型风力发电机设备运行状态检测方法。通过计算提取出的运行状态数据集的多个数据特征,并计算不同特征的权重值,实现数据的多特征融合,并对其进行特征分类。通过计算运行状态数据的偏离阈值和检测阈值,完成对设备运行状态的检测。实验结果表明,和以往的大型风力发电机设备运行状态检测方法相比,设计的基于多特征融合的大型风力发电机设备运行状态检测方法在实际应用中ROC曲线面积较大,检测效果较好。In order to solve the problem of poor effect of traditional large-scale wind turbine equipment operation state detection methods,a large-scale wind turbine equipment operation state detection method based on multi-feature fusion was designed.By calculating the multiple data features of the extracted running state dataset and calculating the weight values of different features,the multi-feature fusion of the data is realized and the features are classified.By calculating the deviation threshold and detection threshold of the operating status data,the detection of the device's operating status was completed.The experimental results showed that compared with previous methods for detecting the operating status of large wind turbine equipment,the designed method based on multi feature fusion for detecting the operating status of large wind turbine equipment had a larger ROC curve area and better detection effect in practical applications.

关 键 词：多特征融合 大型风力发电机设备 运行状态 检测方法 

分 类 号：TM614[电气工程—电力系统及自动化]