volume | PIER Journals

Abstract

In order to improve the efficiency of transformer fault diagnosis and monitoring in power systems, and to realize fault diagnosis of unmanned remote adaptive transformer equipment, we present a method of multi-sensor and multi-direction optical image integrated monitoring in this paper. By monitoring and collecting transformer fault information combined with the changing characteristics of transformer temperature and electrical signals, we establish a transformer calculation model based on multi-level fault and multi-characteristic parameters. According to the characteristics of transformer faults, we use a deep belief network identification (DBNI) algorithm for the transformer and construct the training samples of the transformer diagnosis model using an optimum weight fusion algorithm. The experimental results show that the DBNI model can fully explore the characteristics of large samples, analyze multiple faults information, and extract the hidden features of fault samples. The DBNI model has higher fault diagnosis accuracy than a BP neural network and a single DBN without data fusion and SVM. The DBNI's fault diagnosis accuracy reaching 99.45%. The experimental results show that this model has good robustness of interference ability and can be used intuitively to carry out remote on-line unattended transformer fault diagnosis and information feedback.

1. Dai, H. S., G. Sheng, and X. Jiang, "Dissolved gas analysis of insulating oil for power transformer fault diagnosis with deep belief network," IEEE Transactions on Dielectrics and Electrical Insulation, Vol. 24, No. 5, 2828-2835, Oct. 2017, doi: 10.1109/TDEI.2017.006727.
doi:10.1109/TDEI.2017.006727 Google Scholar

2. Kaur, A., Y. S. Brar, and G. Leena, "Fault detection in power transformers using random neural networks," International Journal of Electrical and Computer Engineering (IJECE), Vol. 9, No. 1, 78-84, February 2019.
doi:10.11591/ijece.v9i1.pp78-84 Google Scholar

3. Mehdipourpicha, H., R. Bo, H. Chen, Md M. Rana, J. Huang, and F. Hu, "Transformer fault diagnosis using deep neural network," 2019 IEEE Innovative Smart Grid Technologies - Asia (ISGT Asia), 4241-4245, 2019, 10.1109/ISGT-Asia.2019.8881052.
doi:10.1109/ISGT-Asia.2019.8881052 Google Scholar

4. Cabanas, M. F., F. Pedrayes González, M. G. Melero, C. H. Rojas García, G. A. Orcajo, J. M. Cano Rodríguez, and J. G. Norniella, "Insulation fault diagnosis in high voltage power transformers by means of leakage flux analysis," Progress In Electromagnetics Research, Vol. 114, 211-234, 2011.
doi:10.2528/PIER11010302 Google Scholar

5. Sun, Z. and Q. Cui, "The application of multiclass support vector machine in power transformer fault diagnosis," Journal of Electronic & Information Technology, No. 10, 25-28, 2019. Google Scholar

6. Mou, S. and T. Xu, "Fault diagnosis method of transformer based on adaptive deep learning model," Journal of Software, Vol. 12, No. 10, 14-18, 2018. Google Scholar

7. Zhang, X. and H. Li, "Research on transformer fault diagnosis method and calculation model by using fuzzy data fusion in multi-sensor detection system," Optik, Vol. 176, 716-723, January 2019. Google Scholar

8. Sun, Z. and L. Xue, "Marginal fisher feature extraction algorithm based on deep learning," Journal of Electronics & Information Technology, Vol. 35, No. 4, 805-811, 2013.
doi:10.3724/SP.J.1146.2012.00949 Google Scholar

9. Jiang, Y. and L. Huang, "Transformer internal fault diagnosis based on DGA and deep belief network," Engineering Journal of Wuhan University, Vol. 50, No. 5, 749-752, 2017. Google Scholar

10. Wang, L. and Y. Zhu, "Parallel phase resolved partial discharge analysis for pattern recognition on massive PD data," Proceeding of the CSEE, Vol. 36, No. 5, 1236-1244, 2016. Google Scholar

11. Zheng, Y. and Y. Zhu, "VPMCD method based on support vector regression and its application in partial discharge pattern recognition," Journal of North China Electric Power University, 2018. Google Scholar

12. Wang, X. and Y. Zhu, "On identification method of power capacitor dielectric loss angle based on deep learning," Transaction of China Electrotechnial Society, Vol. 32, No. 15, 145-150, 2017. Google Scholar

13. Shi, X. and Y. Zhu, "Application of deep learning neural network in fault diagnosis of power transformer," Electric Power Construction, Vol. 36, No. 12, 116-121, 2015. Google Scholar

14. Wang, L. and Y. Xie, "A fault diagnosis method for asynchronous motor using deep learning," Journal of Xi'an Jiaotong University, Vol. 51, No. 10, 128-134, 2017. Google Scholar

15. Geng, L. and X. Liang, "Real-time driver fatigue detection based on morphology infrared features and deep learning," Infrared and Laser Engineering, Vol. 47, No. 2, 2-8, 2018. Google Scholar

16. Zhao, Q. and Z. Li, "Deep multi-task learning for hierarchical classification," Journal of Computer-aided Design & Computer Graphics, Vol. 30, No. 5, 886-891, 2018. Google Scholar

17. Che, C. and H. Wang, "Fault fusion diagnosis of aero-engine based on deep learning," Journal of Beijing University of Aeronautics and Astronauties, Vol. 44, No. 3, 621-627, 2018. Google Scholar

Dr. Xuewei Zhang

Dr. Hanshan Li

Dr. Liping Lu

Dr. Xiaojuan Sun