模块化五电平逆变器子模块开路故障的智能诊断方法
|
摘要
基于深度学习理论,提出了一种基于栈式稀疏自动编码器(SSAE)的模块化五电平逆变器(MFLI)子模块开路故障诊断方法。该方法将MFLI子模块开路故障检测与定位问题转化成分类问题,首先将子模块电容电压信号组合成24通道序列信号,然后沿着24通道序列移动大小为24×40滑动窗口获得"数据带"样本,紧接着将"数据带"转化为向量输入到SSAE中进行逐层无监督特征学习,构建原始故障数据集的深层特征简明表达,最后将深层特征简明表达连接到Softmax分类器输出故障诊断结果。此外,为了提高该方法的抗噪性能,利用已添加高斯白噪声的数据对SSAE进行训练,以提高其特征表达的鲁棒性。结果表明,所提出的故障诊断方法平均准确度达到98.09%,故障平均诊断时间为31.47ms,且具有较高的鲁棒性。
Based on the deep learning theory,a novel method for sub-modular(SM)open-circuit fault diagnosis of modular fivelevel inverter(MFLI)is presented based on the stacked sparse auto-encoder(SSAE).The SM open-circuit fault detection and location problem of MFLI is converted into a classification problem.Firstly,the capacitor voltage signals of all SMs in the MFLI circuit are combined into a 24-channel signal.Then,by moving window along the 24-channel signal with the sliding window,a set of signal segments are acquired which are flattened into vectors and used as SSAE's input subsequently to realize the unsupervised feature learning layer by layer.The deep feature with concise expression of original fault dataset is established and connected to the Softmax classifier to output the fault diagnostic result.In addition,in order to enhance the anti-noise performance of the proposed method,the SSAE is trained by adding Gauss noise to improve the robustness of feature expression.The results show that the proposed fault diagnosis method has the high robustness and versatility with the average accuracy of 98.09% and the average fault diagnosis time of 31.47 ms.
Based on the deep learning theory,a novel method for sub-modular(SM)open-circuit fault diagnosis of modular fivelevel inverter(MFLI)is presented based on the stacked sparse auto-encoder(SSAE).The SM open-circuit fault detection and location problem of MFLI is converted into a classification problem.Firstly,the capacitor voltage signals of all SMs in the MFLI circuit are combined into a 24-channel signal.Then,by moving window along the 24-channel signal with the sliding window,a set of signal segments are acquired which are flattened into vectors and used as SSAE's input subsequently to realize the unsupervised feature learning layer by layer.The deep feature with concise expression of original fault dataset is established and connected to the Softmax classifier to output the fault diagnostic result.In addition,in order to enhance the anti-noise performance of the proposed method,the SSAE is trained by adding Gauss noise to improve the robustness of feature expression.The results show that the proposed fault diagnosis method has the high robustness and versatility with the average accuracy of 98.09% and the average fault diagnosis time of 31.47 ms.
引文
[1]张正发,李可军,王卓迪,等.兼具开路故障诊断能力的MMC电容电压测量方法[J].电力系统自动化,2017,41(7):114-119.DOI:10.7500/AEPS20160712008.ZHANG Zhengfa,LI Kejun,WANG Zhuodi,et al.Measuring method of MMC capacitor voltage with diagnostic ability of open-circuit fault[J].Automation of Electric Power Systems,2017,41(7):114-119.DOI:10.7500/AEPS20160712008.
[2]郭敬梅,曾德辉,王钢,等.基于辅助电路的MMC-HVDC直流故障处理策略[J].电力系统自动化,2016,40(16):90-97.DOI:10.7500/AEPS20151029001.GUO Jingmei,ZENG Dehui,WANG Gang,et al.Auxiliary equipment based processing strategy for MMC-HVDC DC faults[J].Automation of Electric Power Systems,2016,40(16):90-97.DOI:10.7500/AEPS20151029001.
[3]PEREZ M A,BERNET S,RODRIGUEZ J,et al.Circuit topologies,modeling,control schemes,and applications of modular multilevel converters[J].IEEE Transactions on Power Electronics,2015,30(1):4-17.
[4]胡鹏飞,林志勇,周月宾,等.适用于大规模MMC的分布式控制系统及两级均压策略[J].电力系统自动化,2014,38(11):79-84.DOI:10.7500/AEPS20131022006.HU Pengfei,LIN Zhiyong,ZHOU Yuebin,et al.Distributed control system and two-layer voltage balancing method for largescale modular multilevel converters[J].Automation of Electric Power Systems,2014,38(11):79-84.DOI:10.7500/AEPS20131022006.
[5]张建坡,赵成勇,郭丽.模块化多电平换流器子模块拓扑仿真分析[J].电力系统自动化,2015,39(2):106-111.DOI:10.7500/AEPS20130821005.ZHANG Jianpo,ZHAO Chengyong,GUO Li.Simulation analysis on submodule topology of modular multilevel converter[J].Automation of Electric Power Systems,2015,39(2):106-111.DOI:10.7500/AEPS20130821005.
[6]YANG Q,QIN J,SAEEDIFARD M.Analysis,detection,and location of open-switch submodule failures in a modular multilevel converter[J].IEEE Transactions on Power Delivery,2016,31(1):155-164.
[7]DENG F,CHEN Z,KHAN M R,et al.Fault detection and localization method for modular multilevel converters[J].IEEE Transactions on Power Electronics,2014,30(5):2721-2732.
[8]李探,赵成勇,李路遥,等.MMC-HVDC子模块故障诊断与就地保护策略[J].中国电机工程学报,2014,34(10):1641-1649.LI Tan,ZHAO Chengyong,LI Luyao,et al.Sub-module fault diagnosis and the local protection scheme for MMC-HVDC system[J].Proceedings of the CSEE,2014,34(10):1641-1649.
[9]PEUGET R,COURTINE S,ROGNON J P.Fault detection and isolation on a PWM inverter by knowledge-based model[J].IEEE Transactions on Industry Applications,1997,34(6):1318-1326.
[10]CHOWDHURY F N,ARAVENA J L.A modular methodology for fast fault detection and classification in power systems[J].IEEE Transactions on Control Systems Technology,1998,6(5):623-634.
[11]TURPIN C,BAUDESSON P,RICHARDEAU F,et al.Fault management of multicell converters[J].IEEE Transactions on Industrial Electronics,2002,49(5):988-997.
[12]龙伯华,谭阳红,许慧,等.基于量子神经网络的电力电子电路故障诊断[J].电工技术学报,2009,24(10):170-175.LONG Bohua,TAN Yanghong,XU Hui,et al.Fault diagnosis of power electronic circuits based on quantum neural network[J].Transactions of China Electrotechnical Society,2009,24(10):170-175.
[13]WANG T,XU H,HAN J,et al.Cascaded H-bridge multilevel inverter system fault diagnosis using a PCA and multi-class relevance vector machine approach[J].IEEE Transactions on Power Electronics,2015,30(12):7006-7018.
[14]李宁,李颖晖,朱喜华,等.基于混合逻辑动态模型和故障事件识别向量的新型逆变电路故障诊断[J].电网技术,2013,37(10):2808-2813.LI Ning,LI Yinghui,ZHU Xihua,et al.Fault diagnosis for new inverter circuits based on mixed logic dynamic model and incident identification vector[J].Power System Technology,2013,37(10):2808-2813.
[15]陈如清,李强.一种基于频谱分析的可控整流电路故障诊断方法[J].电机与控制学报,2005,9(1):80-85.CHEN Ruqing,LI Qiang.A method for fault diagnosis in controlled rectifier based on frequency analysis[J].Electric Machines and Control,2005,9(1):80-85.
[16]崔江,王强,龚春英.结合小波与Concordia变换的逆变器功率管故障诊断技术研究[J].中国电机工程学报,2015,35(12):3110-3116.CUI Jiang,WANG Qiang,GONG Chunying.Inverter power switch fault diagnosis technique research based on wavelet and Concordia transform[J].Proceedings of the CSEE,2015,35(12):3110-3116.
[17]郑连清,邹涛,娄洪立.电力电子主电路故障诊断方法研究[J].高电压技术,2006,32(3):84-86.ZHENG Lianqing,ZOU Tao,LOU Hongli.Fault diagnosis of power electronic circuits based on wavelet transform and neural network[J].High Voltage Engineering,2006,32(3):84-86.
[18]胡清,王荣杰,詹宜巨.基于支持向量机的电力电子电路故障诊断技术[J].中国电机工程学报,2008,28(12):107-111.HU Qing,WANG Rongjie,ZHAN Yiju.Fault diagnosis technology based on SVM in power electronics circuit[J].Proceedings of the CSEE,2008,28(12):107-111.
[19]DAHL G E,YU D,DENG L,et al.Context-dependent pretrained deep neural networks for large-vocabulary speech recognition[J].IEEE Transactions on Audio Speech&Language Processing,2012,20(1):30-42.
[20]HANNUN A,CASE C,CASPER J,et al.Deep speech:scaling up end-to-end speech recognition[EB/OL].[2014-12-19].https://arxiv.org/obs/1412.5567.
[21]CHEN K,HU J,HE J.A framework for automatically extracting overvoltage features based on sparse autoencoder[J].IEEE Transactions on Smart Grid,2018,9(2):594-604.
[22]徐政,屠卿瑞,管敏渊.柔性直流输电系统[M].北京:机械工业出版社,2013.
[23]李可军,张正发,刘智杰,等.MMC桥臂不对称运行特性分析及子模块故障下的控制策略[J].高电压技术,2016,42(10):3059-3067.LI Kejun,ZHANG Zhengfa,LIU Zhijie,et al.Characteristics analysis of MMC in arm asymmetrical operation and control strategy under submodule fault[J].High Voltage Engineering,2016,42(10):3059-3067.
[24]管敏渊,徐政.模块化多电平换流器子模块故障特性和冗余保护[J].电力系统自动化,2011,35(16):94-98.GUAN Minyuan,XU Zheng.Redundancy protection for submodule faults in modular multilevel converter[J].Automation of Electric Power Systems,2011,35(16):94-98.
[25]TAO C,PAN H,LI Y,et al.Unsupervised spectral-spatial feature learning with stacked sparse autoencoder for hyperspectral imagery classification[J].IEEE Geoscience&Remote Sensing Letters,2015,12(12):2438-2442.
[26]XU J,XIANG L,LIU Q,et al.Stacked sparse autoencoder(SSAE)for nuclei detection on breast cancer histopathology images[J].IEEE Transactions on Medical Imaging,2016,35(1):119-130.
[27]代杰杰,宋辉,杨祎,等.基于栈式降噪自编码器的输变电设备状态数据清洗方法[J].电力系统自动化,2017,41(12):224-230.DOI:10.7500/AEPS20161201003.DAI Jiejie,SONG Hui,YANG Yi,et al.Cleaning method for status data of power transmission and transformation equipment based on stacked denoising autoencoders[J].Automation of Electric Power Systems,2017,41(12):224-230.DOI:10.7500/AEPS20161201003.
[28]戴晓爱,郭守恒,任淯,等.基于堆栈式稀疏自编码器的高光谱影像分类[J].电子科技大学学报,2016,45(3):382-386.DAI Xiaoai,GUO Shouheng,REN Yu,et al.Hyperspectral remote sensing image classification using the stacked sparse autoencoder[J].Journal of University of Electronic Science and Technology of China,2016,45(3):382-386.
[29]GLOROT X,BORDES A,BENGIO Y.Deep sparse rectifier neural networks[C]//International Conference on Artificial Intelligence and Statistics,April 11-13,2011,Fort Lorderdale,USA.
[30]ZHU C,BYRD R H,LU P,et al.Algorithm 778:L-BFGSB:Fortran subroutines for large-scale bound-constrained optimization[J].ACM Transactions on Mathematical Software,1997,23(4):550-560.
[31]SHAO S,WATSON A J,CLARE J C,et al.Robustness analysis and experimental validation of a fault detection and isolation method for the modular multilevel converter[J].IEEE Transactions on Power Electronics,2016,31(5):3794-3805.
[2]郭敬梅,曾德辉,王钢,等.基于辅助电路的MMC-HVDC直流故障处理策略[J].电力系统自动化,2016,40(16):90-97.DOI:10.7500/AEPS20151029001.GUO Jingmei,ZENG Dehui,WANG Gang,et al.Auxiliary equipment based processing strategy for MMC-HVDC DC faults[J].Automation of Electric Power Systems,2016,40(16):90-97.DOI:10.7500/AEPS20151029001.
[3]PEREZ M A,BERNET S,RODRIGUEZ J,et al.Circuit topologies,modeling,control schemes,and applications of modular multilevel converters[J].IEEE Transactions on Power Electronics,2015,30(1):4-17.
[4]胡鹏飞,林志勇,周月宾,等.适用于大规模MMC的分布式控制系统及两级均压策略[J].电力系统自动化,2014,38(11):79-84.DOI:10.7500/AEPS20131022006.HU Pengfei,LIN Zhiyong,ZHOU Yuebin,et al.Distributed control system and two-layer voltage balancing method for largescale modular multilevel converters[J].Automation of Electric Power Systems,2014,38(11):79-84.DOI:10.7500/AEPS20131022006.
[5]张建坡,赵成勇,郭丽.模块化多电平换流器子模块拓扑仿真分析[J].电力系统自动化,2015,39(2):106-111.DOI:10.7500/AEPS20130821005.ZHANG Jianpo,ZHAO Chengyong,GUO Li.Simulation analysis on submodule topology of modular multilevel converter[J].Automation of Electric Power Systems,2015,39(2):106-111.DOI:10.7500/AEPS20130821005.
[6]YANG Q,QIN J,SAEEDIFARD M.Analysis,detection,and location of open-switch submodule failures in a modular multilevel converter[J].IEEE Transactions on Power Delivery,2016,31(1):155-164.
[7]DENG F,CHEN Z,KHAN M R,et al.Fault detection and localization method for modular multilevel converters[J].IEEE Transactions on Power Electronics,2014,30(5):2721-2732.
[8]李探,赵成勇,李路遥,等.MMC-HVDC子模块故障诊断与就地保护策略[J].中国电机工程学报,2014,34(10):1641-1649.LI Tan,ZHAO Chengyong,LI Luyao,et al.Sub-module fault diagnosis and the local protection scheme for MMC-HVDC system[J].Proceedings of the CSEE,2014,34(10):1641-1649.
[9]PEUGET R,COURTINE S,ROGNON J P.Fault detection and isolation on a PWM inverter by knowledge-based model[J].IEEE Transactions on Industry Applications,1997,34(6):1318-1326.
[10]CHOWDHURY F N,ARAVENA J L.A modular methodology for fast fault detection and classification in power systems[J].IEEE Transactions on Control Systems Technology,1998,6(5):623-634.
[11]TURPIN C,BAUDESSON P,RICHARDEAU F,et al.Fault management of multicell converters[J].IEEE Transactions on Industrial Electronics,2002,49(5):988-997.
[12]龙伯华,谭阳红,许慧,等.基于量子神经网络的电力电子电路故障诊断[J].电工技术学报,2009,24(10):170-175.LONG Bohua,TAN Yanghong,XU Hui,et al.Fault diagnosis of power electronic circuits based on quantum neural network[J].Transactions of China Electrotechnical Society,2009,24(10):170-175.
[13]WANG T,XU H,HAN J,et al.Cascaded H-bridge multilevel inverter system fault diagnosis using a PCA and multi-class relevance vector machine approach[J].IEEE Transactions on Power Electronics,2015,30(12):7006-7018.
[14]李宁,李颖晖,朱喜华,等.基于混合逻辑动态模型和故障事件识别向量的新型逆变电路故障诊断[J].电网技术,2013,37(10):2808-2813.LI Ning,LI Yinghui,ZHU Xihua,et al.Fault diagnosis for new inverter circuits based on mixed logic dynamic model and incident identification vector[J].Power System Technology,2013,37(10):2808-2813.
[15]陈如清,李强.一种基于频谱分析的可控整流电路故障诊断方法[J].电机与控制学报,2005,9(1):80-85.CHEN Ruqing,LI Qiang.A method for fault diagnosis in controlled rectifier based on frequency analysis[J].Electric Machines and Control,2005,9(1):80-85.
[16]崔江,王强,龚春英.结合小波与Concordia变换的逆变器功率管故障诊断技术研究[J].中国电机工程学报,2015,35(12):3110-3116.CUI Jiang,WANG Qiang,GONG Chunying.Inverter power switch fault diagnosis technique research based on wavelet and Concordia transform[J].Proceedings of the CSEE,2015,35(12):3110-3116.
[17]郑连清,邹涛,娄洪立.电力电子主电路故障诊断方法研究[J].高电压技术,2006,32(3):84-86.ZHENG Lianqing,ZOU Tao,LOU Hongli.Fault diagnosis of power electronic circuits based on wavelet transform and neural network[J].High Voltage Engineering,2006,32(3):84-86.
[18]胡清,王荣杰,詹宜巨.基于支持向量机的电力电子电路故障诊断技术[J].中国电机工程学报,2008,28(12):107-111.HU Qing,WANG Rongjie,ZHAN Yiju.Fault diagnosis technology based on SVM in power electronics circuit[J].Proceedings of the CSEE,2008,28(12):107-111.
[19]DAHL G E,YU D,DENG L,et al.Context-dependent pretrained deep neural networks for large-vocabulary speech recognition[J].IEEE Transactions on Audio Speech&Language Processing,2012,20(1):30-42.
[20]HANNUN A,CASE C,CASPER J,et al.Deep speech:scaling up end-to-end speech recognition[EB/OL].[2014-12-19].https://arxiv.org/obs/1412.5567.
[21]CHEN K,HU J,HE J.A framework for automatically extracting overvoltage features based on sparse autoencoder[J].IEEE Transactions on Smart Grid,2018,9(2):594-604.
[22]徐政,屠卿瑞,管敏渊.柔性直流输电系统[M].北京:机械工业出版社,2013.
[23]李可军,张正发,刘智杰,等.MMC桥臂不对称运行特性分析及子模块故障下的控制策略[J].高电压技术,2016,42(10):3059-3067.LI Kejun,ZHANG Zhengfa,LIU Zhijie,et al.Characteristics analysis of MMC in arm asymmetrical operation and control strategy under submodule fault[J].High Voltage Engineering,2016,42(10):3059-3067.
[24]管敏渊,徐政.模块化多电平换流器子模块故障特性和冗余保护[J].电力系统自动化,2011,35(16):94-98.GUAN Minyuan,XU Zheng.Redundancy protection for submodule faults in modular multilevel converter[J].Automation of Electric Power Systems,2011,35(16):94-98.
[25]TAO C,PAN H,LI Y,et al.Unsupervised spectral-spatial feature learning with stacked sparse autoencoder for hyperspectral imagery classification[J].IEEE Geoscience&Remote Sensing Letters,2015,12(12):2438-2442.
[26]XU J,XIANG L,LIU Q,et al.Stacked sparse autoencoder(SSAE)for nuclei detection on breast cancer histopathology images[J].IEEE Transactions on Medical Imaging,2016,35(1):119-130.
[27]代杰杰,宋辉,杨祎,等.基于栈式降噪自编码器的输变电设备状态数据清洗方法[J].电力系统自动化,2017,41(12):224-230.DOI:10.7500/AEPS20161201003.DAI Jiejie,SONG Hui,YANG Yi,et al.Cleaning method for status data of power transmission and transformation equipment based on stacked denoising autoencoders[J].Automation of Electric Power Systems,2017,41(12):224-230.DOI:10.7500/AEPS20161201003.
[28]戴晓爱,郭守恒,任淯,等.基于堆栈式稀疏自编码器的高光谱影像分类[J].电子科技大学学报,2016,45(3):382-386.DAI Xiaoai,GUO Shouheng,REN Yu,et al.Hyperspectral remote sensing image classification using the stacked sparse autoencoder[J].Journal of University of Electronic Science and Technology of China,2016,45(3):382-386.
[29]GLOROT X,BORDES A,BENGIO Y.Deep sparse rectifier neural networks[C]//International Conference on Artificial Intelligence and Statistics,April 11-13,2011,Fort Lorderdale,USA.
[30]ZHU C,BYRD R H,LU P,et al.Algorithm 778:L-BFGSB:Fortran subroutines for large-scale bound-constrained optimization[J].ACM Transactions on Mathematical Software,1997,23(4):550-560.
[31]SHAO S,WATSON A J,CLARE J C,et al.Robustness analysis and experimental validation of a fault detection and isolation method for the modular multilevel converter[J].IEEE Transactions on Power Electronics,2016,31(5):3794-3805.