摘要
为了提高对永磁同步风力发电机GSC开路故障的诊断准确率,提出一种对永磁同步风力发电机网侧变流器(GSC)单一和双开路故障进行诊断的新方法。首先,对永磁同步风力发电机GSC的22种开路故障状态(包含1种正常状态)下的三相电流波形进行采样,并作为原始信号数据;之后利用变分模态分解(VMD)将原始信号数据进行分解,并将得到的各层模态系数进行故障趋势特征分析,得到故障特征向量;最后,将特征向量输入深度置信网络(DBN)进行训练和决策,得到分类结果。实验结果表明,所提方法可以对网侧功率变流器的单一和双开路故障进行诊断,且准确率高于其他对比方法。
In order to improve the accuracy of open circuit fault diagnosis of the grid side converter(GSC)of PMSWG(permanent⁃magnet synchronous wind generator),a new method of single and double open circuit fault diagnosis of GSC of PMSWG is proposed.The three⁃phase current waveforms of the GSC of PMSWG under 22 kinds of open circuit fault states(including 1 normal state)are sampled and used as original signal data.After that,the original signal data is decomposed by variational mode decomposition(VMD),and the obtained modal coefficients of each layer are analyzed for fault trend feature to obtain the fault feature vector.Finally,the feature vector is input into the deep belief network(DBN)for training and decision making and obtain the classification results.The experimental results show that the proposed method can diagnose single and double open circuit faults of grid side power converter,and its accuracy is higher than other contrastive methods.
作者
张靖轩
孙鹤旭
孙泽贤
龚思远
董维超
ZHANG Jingxuan;SUN Hexu;SUN Zexian;GONG Siyuan;DONG Weichao(School of Artificial Intelligence,Hebei University of Technology,Tianjin 300130,China;College of Electrical Engineering,North China University of Science and Technology,Tangshan 063210,China;School of Electrical Engineering,Hebei University of Science and Technology,Shijiazhuang 050018,China;Department of Science&Information Technology,Hebei Construction&Investment Group Co.,Ltd.,Shijiazhuang 050000,China)
出处
《现代电子技术》
2021年第3期134-139,共6页
Modern Electronics Technique
基金
河北省教育厅科技项目(QN16214510D)
河北省教育厅科学技术研究项目资助(JQN2020020)
河北省自然科学基金重点项目(E2018210044)。
