期刊文献+

基于CNN-LSTM的风力发电机组轴承异常检测 被引量:6

Abnormal Bearing Detection of Wind Turbine Based on CNN-LSTM
在线阅读 下载PDF
导出
摘要 为保障风力发电机组的可靠运行,基于数据的机组异常状态检测尤为重要。文章提出一种基于级联深度学习模型的风力发电机组主轴承异常状态检测方法,首先利用风力发电机组机理知识和数据间的相关性选择与主轴承密切相关的参数,然后建立基于卷积神经网络(CNN)和长短记忆网络(LSTM)的观测参数与目标参数的逻辑关系,并且通过均方根误差评估模型预测温度与实际采集温度的差异。最后通过海上某风电场SCADA数据进行算例验证。结果表明:CNN-LSTM模型不仅能够更早得发现主轴承异常状态,还能够发现LSTM发现不了的主轴承异常特征。 In order to ensure the reliable operation of wind turbine,the abnormal state detection of wind turbine based on data is particularly important.An abnormal state detection method of wind turbine bearing based on cascade deep learning model is proposed.Firstly,the parameters closely related to the bearing are selected by using the correlation between wind turbine mechanism knowledge and data,and then the logical relationship between observation parameters and target parameters based on convolutional neural network(CNN)and long short memory network(LSTM)is established.The difference between the predicted temperature and the actual collected temperature is evaluated by the root mean square error evaluation model.Finally,an example is verified by the SCADA data of a wind farm in Northwest China.The results show that the CNN-LSTM model can find the abnormal state of the bearing earlier,and the CNN-LSTM model can find the abnormal characteristics of the bearing that the LSTM cannot find.
作者 温钊 冉军 张宁 刘冰冰 胡号朋 周庆梅 WEN Zhao;RAN Jun;ZHANG Ning;LIU Bingbing;HU Haopeng;ZHOU Qingmei(CSIC Haizhuang Windpower Co.,Ltd.,Chongqing 401120,China;Southern Marine Science and Engineering Guangdong Laboratory(Zhanjiang),Zhanjiang 524013,Guangdong,China)
出处 《船舶工程》 CSCD 北大核心 2022年第S02期16-19,共4页 Ship Engineering
基金 国家重点研发计划(2020YFB1506600) 国家自然科学基金(U2141245) 重庆市科技局(ZC22.004)
关键词 风力发电机组 轴承 深度学习 CNN-LSTM wind turbine bearing deep learning CNN-LSTM
作者简介 温钊(1992—),男,硕士、工程师。研究方向:大数据建模。
  • 相关文献

参考文献6

二级参考文献52

共引文献77

同被引文献110

引证文献6

二级引证文献22

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部