Intelligent Localization of Transformer Internal Degradations Combining Deep Convolutional Neural Networks and Image Segmentation

Industrial 4.0 placed higher demands on the field of intelligent equipment monitoring. The transformer is one of the critical power devices, its intelligent monitoring and fault positioning require in-depth studies. In this study, an efficient fault localization method for transformer internal thermal faults was proposed by introducing different deep convolutional neural networks (CNNs) and image segmentation. First, the transformer monitoring images of temperature and velocity fields in fault conditions were simulated using the lattice Boltzmann method (LBM), and the images were also used to highlight features information. In practice, transformer degradation does not frequently occur, so that the fault samples for deep learning are insufficient. To solve this problem, a transfer learning method was employed. Subsequently, fault locations were defined as classification labels, and different CNN’s were used to classify the labels to achieve the fault localization results. Next, image segmentation was performed to extract the features of fault areas and simplify the data volumes. Likewise, the CNN’s were employed to perform the fault localization again. Afterward, since the monitoring sensors were not located everywhere in a transformer in practical applications, information of partial monitoring areas where the monitoring sensors located was trained following a similar procedure. After image segmentation, the average fault localization accuracy using the information obtained by sensors decreased from 97.95% to 94.42%, while the data volume was reduced to nearly 1% of the original one. Besides, the average calculation time per iteration decreased by 8.816%, while the loss value was reduced by 37.68%. Finally, the Friedman hypothesis test and Nemenyi post hoc test were performed to compare the evaluation indicators of different networks, and the performance of GoogLeNet in this case was considered the best.

[1]  Xinghao Ding,et al.  Vehicle Type Recognition in Surveillance Images From Labeled Web-Nature Data Using Deep Transfer Learning , 2018, IEEE Transactions on Intelligent Transportation Systems.

[2]  Pierre Beauseroy,et al.  Transfer learning for one class SVM adaptation to limited data distribution change , 2017, Pattern Recognit. Lett..

[3]  Lixiang Duan,et al.  Segmented infrared image analysis for rotating machinery fault diagnosis , 2016 .

[4]  Min Wu,et al.  An overview of state-of-the-art partial discharge analysis techniques for condition monitoring , 2015, IEEE Electrical Insulation Magazine.

[5]  Ibrahim Senol,et al.  Development of a hot-spot temperature calculation method for the loss of life estimation of an ONAN distribution transformer , 2018 .

[6]  Dumitru Erhan,et al.  Going deeper with convolutions , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[7]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[8]  Rik Van de Walle,et al.  Deep Learning for Infrared Thermal Image Based Machine Health Monitoring , 2017, IEEE/ASME Transactions on Mechatronics.

[9]  Bo Li,et al.  LSTM-Based Analysis of Industrial IoT Equipment , 2018, IEEE Access.

[10]  Jawad Faiz,et al.  Dissolved gas analysis evaluation in electric power transformers using conventional methods a review , 2017, IEEE Transactions on Dielectrics and Electrical Insulation.

[11]  A. Golneshan,et al.  Numerical Study of Temperature Distribution in a Spherical Tissue in Magnetic Fluid Hyperthermia Using Lattice Boltzmann Method , 2011, IEEE Transactions on NanoBioscience.

[12]  Lin Zhao,et al.  A Deep Feature Optimization Fusion Method for Extracting Bearing Degradation Features , 2018, IEEE Access.

[13]  Mingming Zhang,et al.  Identification of Maize Leaf Diseases Using Improved Deep Convolutional Neural Networks , 2018, IEEE Access.

[14]  Fengshou Gu,et al.  Thermal image enhancement using bi-dimensional empirical mode decomposition in combination with relevance vector machine for rotating machinery fault diagnosis , 2013 .

[15]  Bin Wang,et al.  A Fast and Robust Level Set Method for Image Segmentation Using Fuzzy Clustering and Lattice Boltzmann Method , 2013, IEEE Transactions on Cybernetics.

[16]  Kilian Q. Weinberger,et al.  Densely Connected Convolutional Networks , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[17]  Changqing Shen,et al.  Stacked Sparse Autoencoder-Based Deep Network for Fault Diagnosis of Rotating Machinery , 2017, IEEE Access.

[18]  Gerasimos Rigatos,et al.  Power transformers’ condition monitoring using neural modeling and the local statistical approach to fault diagnosis , 2016 .

[19]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[20]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[21]  Abdulmajeed A. Mohamad.  The Boltzmann Equation , 2011 .

[22]  Md Mominul Islam,et al.  A review of condition monitoring techniques and diagnostic tests for lifetime estimation of power transformers , 2017, Electrical Engineering.

[23]  Yan-Jun Chen,et al.  Experiment and simulation of natural convection heat transfer of transformer oil under electric field , 2017 .

[24]  Paolo Napoletano,et al.  Benchmark Analysis of Representative Deep Neural Network Architectures , 2018, IEEE Access.

[25]  Xiang Zhang,et al.  Interpretation of Hot Spot Factor for Transformers in OD Cooling Modes , 2018, IEEE Transactions on Power Delivery.

[26]  Juliano R. da Silva,et al.  Online Evaluation of Power Transformer Temperatures Using Magnetic and Thermodynamics Numerical Modeling , 2017, IEEE Transactions on Magnetics.

[27]  Fabrice Ville,et al.  On the use of temperature for online condition monitoring of geared systems - A review , 2018 .

[28]  L. Luo,et al.  Theory of the lattice Boltzmann method: From the Boltzmann equation to the lattice Boltzmann equation , 1997 .

[29]  Kaiming He,et al.  Mask R-CNN , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[30]  Jiangtao Wen,et al.  Intelligent Bearing Fault Diagnosis Method Combining Compressed Data Acquisition and Deep Learning , 2018, IEEE Transactions on Instrumentation and Measurement.

[31]  S. P. Ang,et al.  Causes of transformer failures and diagnostic methods – A review , 2018 .

[32]  Shiyi Chen,et al.  LATTICE BOLTZMANN METHOD FOR FLUID FLOWS , 2001 .

[33]  Stefan Tenbohlen,et al.  Investigation of Thermal Behavior of an Oil-Directed Cooled Transformer Winding , 2018, IEEE Transactions on Power Delivery.

[34]  M. Lehtonen,et al.  Dynamic thermal modeling of power transformers: further Development-part II , 2006, IEEE Transactions on Power Delivery.

[35]  Mehdi Bagheri,et al.  Transformer Fault Condition Prognosis Using Vibration Signals Over Cloud Environment , 2018, IEEE Access.

[36]  Xiaowei Luo,et al.  Transfer learning and deep convolutional neural networks for safety guardrail detection in 2D images , 2018 .

[37]  Kyungjae Lee,et al.  Brightness-Based Convolutional Neural Network for Thermal Image Enhancement , 2017, IEEE Access.

[38]  Forrest N. Iandola,et al.  SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <1MB model size , 2016, ArXiv.

[39]  Seyednaser Hashemnia.  Charactrisation of power transformer frequency response signature using finite element analysis , 2014 .

[40]  T. Mariprasath,et al.  A real time study on condition monitoring of distribution transformer using thermal imager , 2018 .

[41]  R. E. Lee,et al.  Distribution-free multiple comparisons between successive treatments , 1995 .

[42]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[43]  Jie Wang,et al.  Transferring Pre-Trained Deep CNNs for Remote Scene Classification with General Features Learned from Linear PCA Network , 2017, Remote. Sens..